1.1 Difficulties in defining genres

Undoubtedly, the gaming industry has dramatically changed over the years, especially when it comes to game genres and mechanics [5]. In 1980, due to the economic pressure in the gaming industry, games started to be categorized into sets of genres that shared narrative structure, gameplay mechanics, dynamics, and schemes. One of the first video game genre classifications was created by Crawford [6]. “Skill-and-action games”—combat, racing, and maze games—were distinguished from “Strategy games”, which included roleplaying (RPG), adventure, and educational games. Early taxonomies, even if extensive and elaborated as in Wolf’s 42 video game categories [7] or in the typology created by Aarseth et al., [8], could not fit into modern game genres such as Massively Multiplayer Online Role-Playing Game (MMORPG) or Multiplayer Online Battle Arena (MOBA). Another problem was that some taxonomies created by scientists did not fulfill their communicational function and due to that they were not compatible with the market and community [9].

Taxonomical obstacles emerged because of the progress of blurring the lines between game genres. Today’s games are harder to distinguish purely by their characteristic mechanics; for example, Action games mechanics can be found in multiple other genres [5]. Such a state led to the creation of multigenre hybrids such as Action-RPG or Action-adventure. After all, the enormously popular League of Legends combines elements of real-time strategy and action, which creates the MOBA genre. Another problem emerges from the development of games that combine not two but three or more different genres, such as Fortnite or Minecraft, which are even more difficult to classify [5,10]. Heintz et al. [11] emphasize that the relation between the genres started to overlap and that the genres themselves are not consistently defined. Even the sellers categorize video games into multiple genres to increase the possibility of someone buying their product. Furthermore, games can be categorized differently depending on which attributes spark the player’s interest. For example, Genshin Impact can be played for narrative reasons or for the gameplay focused on grinding skills and equipment of the character. Consequently, any given game can be included in one genre on the basis of its gameplay and in another because of its iconography or story [12]. Heintz et al. [11] tried to map the attributes characteristic of extracted game genres–e.g., Adventure includes collecting objects, communicating, emotional connection to character, puzzles, searching, fantasy world, exploration, and story. However, again, it is visible that some of the attributes were shared across multiple genres or even overlapped with each other. Another trend in the gaming industry is creating game genres on the basis of mechanics like “First person” or “Real time strategy” (RTS). The first example was usually connected to Shooters or Action games [4]. On the other hand, RTS and TBS (Turn-based strategy) emerged as new genres from Strategy [13]. Nevertheless, despite these difficulties, psychologists have been trying to classify game genres for decades. This is because game characteristics have repeatedly been shown to be related to psychological phenomena. There is no indication that it will be different in the future; still, understanding the interaction between gamer characteristics, situational conditions, and game features will be the key to understanding and predicting human behavior, thoughts, and emotions.

1.2 Game characteristics and their classification are important subjects of psychological research

There are many areas of psychological research where games’ features have proven to play a role. Among others, these are: aggressive behavior [e.g., 14], education [e.g., 15], health [e.g., 16], sexual behavior [e.g., 17], and many others. Two fields seem to be particularly interesting for researchers in recent years: the relationships between games’ characteristics and gaming disorder and cognitive development. Below, we briefly present examples of research in these areas to illustrate that it is impossible to ignore the features of games in research on human interaction with games.

2.1 Search strategy

This systematic review was conducted according to the Preferred Reporting Items for Systematic Review and Meta-Analyses [PRISMA; 47]. The aim of the study was to capture the most recent and the most prominent methods of classifying video game genres in empirical studies in disciplines such as psychology and computer sciences. The search was conducted on the following databases: ScienceDirect, Taylor and Francis, Scopus, and PubMed. The last search was performed on December 11th, 2023. The results were limited to the filters of subject area: Psychology, Psychiatry, Computer Science and Neuroscience. The search strategy combined key terms such as “video game genre” OR “video game classification” OR “video game categories” OR “videogame genre” OR “videogame classification” OR “videogame categories” OR “digital game genre” OR “digital game classification” OR “digital game categories”.The databases were searched twice systematically and independently by two reviewers. In case of discrepancies, they were resolved through clarification and consensus. Both reviewers collected data from databases that they had screened. Articles were included in the study if they met the following criteria:

1. The articles were published between January 2012 and December 2023 (inclusively). It was due to the enormous changes in the video games market and the need to present actual video game genre classifications;
2. The articles were peer-reviewed and published in English or Polish;
3. The articles had an empirical nature;
4. The articles were related to the science disciplines such as psychology, psychiatry, neuroscience or computer science;
5. The articles included comprehensive video game classification and thus did focus on the comparison more thantwo genres.

The exclusion criteria were:

1. The studies had theoretical nature;
2. The studies focused on one or two genres;
3. The articles includes genres named specifically for the sake of the study (e.g., violent game genres).

Initially, 1229 results were retrieved, including duplicate results. The titles and abstracts of the results from all the searched databases were screened for relevance, which led to the identification of 176 potentially eligible articles, of which 161 were retrieved. After reading entire papers, 64 articles were excluded from the study due to not meeting the criteria of inclusion. Consequently, 93 full-text articles were included in this review. In addition, on the basis of the analysis of references of the identified articles, 4 additional papers were assessed for eligibility, of which 3 were included in the analysis. As a result, 96 papers were analyzed. Fig 1 provides a flowchart outlining the search for the studies.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. PRISMA diagram of the search and selection process for the relevant literature.

https://doi.org/10.1371/journal.pone.0299819.g001

2.2 Data coding/retrieval

The following data were retrieved from all the eligible articles: list of genres, information whether the authors proposed any subgenres (yes/no), information whether the authors provided definitions of genres (yes/no), and information whether the authors provided sample games belonging to genres (yes/no). Furthermore, sources of classification included in the articles were collected to assess their study-uniqueness. We planned to code the source of classification as follows: “Article reference” (when the authors used a taxonomy previously used in another published work), “Website” (when the authors used a taxonomy published on a specialized website), “Game platform” (when the authors used a taxonomy published on a game selling platform), “Arbitrary: authors” (when the taxonomy was made by the authors themselves for this study, without reference to external sources), “Arbitrary: participants” (when the taxonomy was made on a basis of participants responses, without reference to external sources), “Expert judges” (when experts external to the authors provided the taxonomy), “Arbitrary: not specified” (when the taxonomy source was not specified), “Combination” (when the taxonomy used resulted from the compilation of sources of various nature). One additional category (“Market analysis”) emerged during classification when the authors decided to use the market analysis as a basis for the games’ taxonomy. Moreover, information about the country and the number of participants by gender who took part in the studies was collected. In some cases, the subject of empirical analysis was data collected without the direct participation of the respondents, in such a situation the authors did not provide the number of respondents, as a consequence, we also did not provide N, we marked such studies as ’not applicable’. Impact factor of the scientific journal where articles were published was retrieved to assess the possible quality of the work. All these data are presented in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Description of the studies covered in the systematic review.

https://doi.org/10.1371/journal.pone.0299819.t001

3.1 Quality of creating game taxonomies

Gaming taxonomies distinguished by the authors of articles included in this systematic review have multiple different sources. During our review, we identified 46 unique taxonomies in 96 papers. It means that approximately 48% of taxonomies were used only once. Moreover, only in the case of 50 papers (52%) authors utilized some kind of external or combined source of taxonomy. The specific numbers are shown in Fig 2 below.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. The number of empirical articles depending on the source of the game taxonomy used in them.

Notes. Red bars represent study-specific taxonomies (including arbitrary); blue bars represent taxonomies based on external, publicly available, sources.

https://doi.org/10.1371/journal.pone.0299819.g002

Unfortunately, the way of constructing game genre classification was not specified in eighteen (19%) articles [31,45,46,48,58,60–63,67–69,74,76,111,114,121,126]. It can be assumed that the construction process was arbitrary. On the other hand, in twenty-three articles, the classification was based on the assessment of the authors and/or was study-specific [19,20,25,27,51,56,64,70,72,75,78,82,87,93,99–101,104,112,123,124,127,128]. Furthermore, the classification that was based solely on the judgments of the participants appeared in three articles [80,83,113]. Mathews et al. [95] created a game genre taxonomy by using expert judges. Moreover, some researchers used the following websites to gain information about game genre: Gamefaqs.com, Pegi database, Naver database, Gamespot.com IGN.com, Amazon.de, Amazon.jp [33,44,53,59,77,84,96,97,103,108,119]. Except for websites, gaming platforms such as Steam were also used to create game classifications [50,54,73,89]. Gilbert et al. [71] and Rodio & Bastien [110] built their classification by analyzing the market sales in 2010. Some taxonomies (n = 19) were distinguished based on previous literature [29,32,49,52,54,57,64,90–92,102,105,106,117,118,120,125,129,130]. Articles used as references for other taxonomies were as follows: Lucas and Sherry [131], Apperley [43], Zammito et al. [132], Adams [133], Lee et al. [85], Lemmens and Henriks [86], Rehbein et al. [109], Dale and Green [134], Green et al. [72]. Other researchers (n = 12) used various combinations of the abovementioned sources. [30,55,65,79,81,85,86,88,94,98,107,109,115,116,122]; out of these 15 cases, in 12 authors combined previous articles with other sources. The way in which game taxonomies are constructed seems not to be dependent on the journal’s impact factor. However, it is worth mentioning that in journals which impact factor is above 11, authors sourced their game classification on a website, Gamefaqs.com, and also provided definitions for each single genre [66,108]. It is important to mention, thatYilmaz, Yel and Griffiths (129,130) presented three video game taxonomies. The first was focused on game’s scenarios, second primary production purposes and last one presented traditional game genres.

It is also important to mention that only twenty-six articles not only listed several game genres included in their research but also defined them [20,21,44,46,51,53,57,66,71,74,76,78,81,85,89,94,95,98,105,108,112,118–120,125,126]. However, some provided definitions are not separable; e.g., Thorne [119] used “puzzle-solving” phrase to describe both Action/adventure and strategy genres. Examples of games that were categorized into specific genres were provided in almost half of the articles included in the systematic review (n = 37). Usually, the article which contains the definition also presents the examples of games characteristic for the genre, however, the authors of twenty-two papers did not mention neither definition nor examples. The lack or presence of definition and examples of games for various genres seems not to be related to the impact factor of the scientific journal.

3.2 General characteristics of taxonomies

The list of genres included in all of the studies is presented above in Table 1 and in S1 Table in Supporting information. It follows from them that the most mentioned genres in all the articles were Sport (n = 75), RPG (n = 75), Strategy (n = 57), Simulation (n = 56) and Puzzle (n = 55), Driving/Racing (n = 50) MMORPG (n = 44) and FPS (n = 43). As was already mentioned, Sport was the most mentioned genre in the studies included in this systematic review. However, it is important to emphasize that multiple studies not only include broad Sport category but also present multiple sport disciplines not as subcategories but as different game genres such as Driving/Racing (n = 50) or Fighting (n = 43). The cause of these differentiations may be connected to the mechanical and storyline aspects of these genres which include not only simple simulation of sport disciplines but also elements of fiction such as racing in fantasy lands (e.g., Mario Kart) or using inhuman powers in fighting games (e.g., Inhumans, Tekken). Furthermore, it is important to emphasize that some researchers debate whether Sport games, which are simulation of real-life sport should be differentiated from Simulation genre [57,94]. On the other hand, Simulation as a genre is not only associated with traditional sports; it can also be used to describe video games that focus on management or construction (such as Sim City, Planet Zoo, Railway Empire, and Train Simulation), which simulate real-life events or activities. Rehbein et al. [109] even distinguished five different Simulation’s subgenres such as Simulation and construction, Life simulation games, Business simulation games, Sandbox physics games and Flight simulation games.

Aspects of management and construction in Simulation games are strongly related to Strategy games. This genre was mentioned 57 times in the articles included in this study. However, it is important to emphasize that researchers usually diversify its subgenres–RTS (n = 30), TBS (n = 13) and Management (n = 8). Some studies included genres that combine Strategy and RPG elements such as Turn based RPG [52,88,121]. MOBA, one of the most popular game genres among players, was mentioned only 27 times, mostly in articles published after 2015. Similarly to Strategy genre, Action (n = 39) also has been usually differentiated into Action-adventure (n = 39). Moreover, this genre is also represented in such multigenres as Action RPG (n = 8) [52,61,72,79,82,88,104,121] or Action-Puzzle [74]. López-Fernández et al. [92] also included Action-shooter as an individual genre without including Shooter or FPS category in their study.

Action games in the past were usually defined as shooters [4], so creating new genre may create unnecessary difficulties in defining genres. Shooter as a genre was included in 33 articles. However, it is important to mention that in most studies where researchers included Shooters in their genre classification, they did not add FPS to it [25,29,55,80,88,89,104,107,108,112]. Furthermore, FPS was included as a subgenre of Shooters in three studies [19,79,109]. Only three studies mentioned TPS (third person shooter) as a subgenre of shooters [19,74,125]. As presented above, FPS has been much more present in the studies than the broader Shooter genre. It implies that first-person shooter started to be used as a replaceable name for shooters even though it is a much narrower category as a subgenre that cannot include TPS in it. Other genres related to Shooters included in the studies were Arcade/Shooting [81], Hero-shooter which is defined as “MOBA inspired first person shooter” [76], Battle Royale [19,76,87,117], Beat ‘em’ up [61,63,84,93,122] and Top-down shooters [88] which are also known as Shoot ‘em’ up [109].

Recently, new mechanics and multigenres such as Adventure RPG [120] and Text Adventure [74,104] were diversified from the broad category of Adventure games (n = 30). Furthermore, some researchers mention Survival and Horror games as an entirely excluded category where player need to survive in the potentially scary and extreme environment [61,74,88,95,112,117,122]. However, some researchers included such games in the Adventure and Thrillers category [94].

As was mentioned before, Puzzle games were presented as an individual genre in 55 articles. Puzzle games are defined as “games involving matching, logic, deductive reasoning and other” [44]. Hazel et al. [74] differentiated three Puzzle genres: Puzzle Action, which combines elements of Puzzle and Action game in which player solves puzzles by using a first- or third-person avatar; Puzzle Other–where solving logical mysteries happens without an avatar and individual category such as Point and Click which is defined as “The player uses a cursor to interact with a flat environment, usually to solve puzzles and experience a story”. Such definition of Point and Click suggests occurrence of puzzle and adventure aspects in game. Furthermore, Hazel et al. [74] distinguished Hidden Objects which definition implies inclusion in the Point and Click category: “A form of Point and Click game where players locate items in a busy environment”. This category also appears in an article by Rehbein et al. [109]. However, it is presented as a subgenre of a broader category called Brain and Skills, which also included Card Games, Puzzle Games, Board Games, Quiz Games, Skill Games, Fitness Games, Music Games, Party Games and Game Collections. As was said before, definition of Puzzle games usually includes aspect of logic [44]. Despite that, Logic genre was differentiated in two articles as an individual genre [69,120].

On the other hand, Platform Games were mentioned 33 times as a standalone genre. However, Platform games have recently been categorised as a subgenre of Action or Puzzle games, even though they are often understood as one of the first genre of games [135]. Elliot et al. [44] defined platform games as games that involve action of jumping and running (e.g., Mario Bros). San Nicolas Romera et al. [112] added avoiding opponents attacks by using some objects or artifacts. It seems important to mention that Jump’n’Run which definition seems to resemble Platform games was mentioned as its subgenre [135]. It was also listed as an Action-adventure subgenre by Rehbein et al. [109] and an unclassified subgenre in an article by von der Heiden et al. [125]. Dieris-Hirche et al. [63] mentioned it as a standalone category. Platform games were also listed as a merged category with Party games by Mazurek et al. [97]. In relation to that, Party as an individual genre was mentioned in seven articles [20,55,56,60,103,104,106,111,120,122]. Furthermore, another gaming genre that can be played individually or with a group of friends is Rhythm/Music games (n = 27). Similarly, recently, Board or Card games have been played with use of specific applications giving the opportunity to play different games (virtual tables) such as the Tabletop Simulator or individual games such as Magic: The Gathering Arena, Heartstone or Legend of Runaterra (n = 23). Some researchers distinguish Gambling and Casino games from Board/Card games category [44,53,66,95,111].

Another category of games are browser games (n = 9) or games on social media apps (n = 19), which are defined as simple games that could run on a simple platform or social media network [20,126]. It is interesting that Wohn et al. [126] listed some subgenres of social network games (SNG) that usually focus on simulation–farm, pet, town or restaurant simulation. They also differentiated such SNG’s subgenres as Card, Brain, Word and Arcade games. Some researchers also include SNG in the casual game genre, alternatively called mobile or idle, in which games are defined as being designed for short periods of play and having a portable nature [27,65,93,104,120]. It is important to mention that category Other for unclassified genre of the games or mixed nature of the game’s genre was included in sixteen articles [19,44,58,63,66,78,91,108,109,112,114,116–118,125,127].

3.3 One genre, multiple definitions

The present review highlighted the definitional inconsistency of the game genres under study. Due to the lack of convergence in the genre and sub-genre classification of games, it is difficult to clearly state for which of the genres the greatest discrepancy in definitions occurred since the definitions particularly depend on the methods of classification. It is noteworthy that, excluding studies that referred to previous studies in the context of classification, total convergence among the publications subject to this review did not appear for any game genre. There were also numerous instances where no definitions were provided and the genres’ classifications were arbitrary [19,25,27,31,32,45,48,56,58,60–64,67–70,72,75,79,80,82,83,87,93,99–101,104,111,113,114,121–124,127,128], thus leaving many unknowns.

One of game-play virtuosity as the gamer controls every move of the counter that usually can kill and can be killed.” [57], “Include physical challenges and require the greatest classification discrepancies concerned Action games due to the variety of definitions, classification as falling under/combining with other genres, or the lack of their inclusion. In the studies subject to this review, the Action genre has been defined disparately, for instance: "Games that fell into the action category shared a focus on physical enactment to advance in the game, often using violent or dangerous behaviors.” [71], “Action games need skills in hand-eye coordination to complete objectives, which include defeating opponents.” [120]. Moreover, some studies have classified the Action-adventure genre as one [19,30,44,48,49,51,54,56,61,64,66,72,79,80,82,83,86,91,95,97,104,105,111,117,119,120,122–124,129,130],some have found them separate [31,32,45,53,55,60,73,93,98,99,102,106,114,115], and some have classified them as both one genre and two separate genres [29,33,52,84,85,96,109,121] drawing further inaccuracies in the context of classifying this genre as deserving a separate distinction or not.

The greatest definitional uniformity was visible in the case of Sport games, which were most often described as simulations of real-life sports (e.g., football, basketball, etc.) [e.g., 20,44,71,76]. However, there were cases where Sport games were treated as part of the Simulation genre [e.g., 57,112,125] or where Sports and Racing games were classified as one genre [e.g., 30,111,119]. There were also two instances where Sport games were classified as a part of the action genre [e.g., 52,99]. The aforementioned examples serve as evidence that even within the genre characterized by the highest degree of classification homogeneity, inaccuracies still persist.

3.4 The multi-genre misunderstanding

The general range of game genres proposed by the authors also differed noticeably, regardless of the year of publication, which also contributed to the diverse understanding of the genres. A study published in 2012 [44] lists the following 16 genres: MMORPGs, Other RPGs, Action-adventure, First-person shooter (FPS), Other shooter, Sports general, Sports other, Rhythm, Driving, Platformer, Real-time strategy, Other strategy, Puzzle, Board and card games, Gambling, Other; whereas a study published in later years [32] mentions the following 9 genres: Strategy, Puzzle, Sport, Adventure, Role-playing (RPG), Platform, Simulation, Music, Action. Surprisingly, a study published a year later [57] proposes only 4 genres: Action, Role-playing, Simulation, Strategy; and another study [103] proposes 8 genres: Action, Fighting, Music, Party, Platformer, Racing, Simulation, Sports.

The presented classifications of game genres are surprising, considering the fact that the gaming market is constantly expanding, offering new genres that are fusions of other previously existing genres (multi-genres), for instance, MMORPGs. Multi-genre game developers extract the immersive mechanics of two or more game genres and combine them into one product to achieve higher player engagement. This phenomenon was highlighted in the case of MMORPGs, which, as demonstrated in the introductory part of the present review, turn out to be one of the most associated with Gaming Disorder. Therefore, in order to get to the essence of this phenomenon by comparing research results and building a coherent set of information on this subject, the uniformity of game genres in research is substantial. Yet, according to the classifications proposed in the mentioned studies, MMORPGs could be classified as MMORPGs themselves but also as Role-playing and Action simultaneously. This way of classifying becomes problematic given the fact that the Action game genre appears in each of the classifications and only one of them recognizes MMORPGs as such, leading to confusion not only in the classification of MMORPGs but also in the understanding of other genres.

3.5 Steam tags–different approach to game classification

Li and Zhang [89] proposed that game genres, instead of strict taxonomies, should be based on the more flexible approach offered by Wittgenstein’s “family resemblance” theory. Researchers tried to construct their gaming classification by collecting and analyzing user-generated tags from the Steam gaming platform. Results of calculated weighted degree, closeness centrality, betweenness centrality, and PageRank show that five game tags obtained the highest scores in these metrics. They were as follows: Indie, Action, Adventure, Singleplayer, and Casual. Furthermore, four game genres were distinguished on the basis of “high-centrality gameplay-designing tags within the model Video Game Metadata Schema category and Steam categorization” [89]. These major game genres were Strategy and Simulation Games (Game Tags’ examples: “Strategy”, “Simulation”, “Multiplayer”, “Survival”, “Sport”, “Medieval”, “Turn-based”) Puzzle and Arcade Games (Game Tags’ examples: “2D” “Great Soundtrack”, “Puzzle”, “Comedy”. “Arcade”, “Platformer”), RPG Games (Game Tags’ examples: “RPG”. “Story Rich”. “Fantasy”, “Choices matter”, “Hack and Slash”, “Visual Novel”, “Point and Click”, “Text-based”) and Shooter Games (Game Tags’ examples: “Atmospheric”, “Sci-fi”, “Shooter”, “Exploration”, “FPS”, “Horror”, “TPS”, “Violent”). Such a way of presenting genres includes major story themes, mechanics or subgenres, ratings, and atmosphere/mood.

Furthermore, Li [88] in his next study conducted Exploratory Factor Analysis (EFA) on game tags from 22 749 video games on the Steam platform. Results of EFA implied the presence of 29 game genre factors: Arcade Shooter, Fighting, Soccer, Sandbox, TBT, Resources Management, Music, Shooter, Rogue, Character Action, Strategy, Classic, Board/Card, Gal Game, Platformer, Survival, MMO, Racing, RPG, Strategy RPG, ARPG (Action RPG), Top-down Shooter, Interactive Fiction, Tower Defence, RTS, Simulation, Exploration, Parkour and Education. Full table with correlated Gameplay Tags can be found in Table 1, page 212 [88]. Even though twenty-nine genres were distinguished by EFA it still did not create totally separable game genres. Furthermore, Li [88] also noticed that the new generation of video games is multigenre by nature. In turn, analyzed games belonged to one single genre or a combination of several genres. Li suggested that each game can be characterized by a certain list of game genre factors; for example, Divinity Original Sin 2 can be described as “Turn Based Tactics” and “RPG”. Using game factors may not only facilitate customer needs but also be important for scientific reasons.

4.1 Yet another old chestnut

This systematic review summarizes video game classifications used in psychological research between 2012 and 2023. We propose to create a game genre classification that consists of all the video game genres mentioned in the coded articles. As presented in S1 Table in Supporting information, the video game genres listed from the most mentioned are as follows: Sport, RPG, Strategy, Simulation, Puzzle, Driving/Racing, MMORPG, FPS, Fighting, Action-adventure, Action, Platform, Shooter, Rhythm/Music, MOBA, Arcade, TBS, Educational, Browser, Party, Management, Action RPG, Survival/Horror, Sandbox, Beat ‘em’ up.p. Genres mentioned less than five times are recorded in the section Other, e.g., Gambling, TPS, Retro, Logic, Battle Royale, Hero-Shooting, Quiz/Trivia, Roguelike etc. This categorization presents over a decade of research regarding using video games genres and it can be useful material for further studies if we decide to continue using video game genres. However, it must be remembered that due to the fast-paced development of the gaming industry and the creation of new game genres that combine the mechanics and characteristics of two or more genres, it may not stand the test of time in the long run.

4.2 New approach

The essence of psychological research is to understand and explain human behavior, emotions, and cognitive processes. It is this focus on understanding their mechanisms at the individual, interpersonal, and intergroup levels that distinguishes psychology from other disciplines that may study the same phenomena but take different perspectives. With this in mind, the question arises: “What is the purpose of investigating games’ characteristics in the context of psychology?”. There is no doubt that the first researchers dealing with this topic set themselves the goal of understanding the relationship between psychological phenomena and the characteristics of games. For instance, McClure and Mears [138] attempted to answer the question of the relationship between ’intellectual efficiency’ and video game genre preferences. They report the following discovery: "Brighter people (. . .), liked simulation and adventure games more than arcade games" (p. 274). The authors do not explore this discovery in the discussion of the results, but a certain distinction is striking—simulation and adventure games seem to be more complex than arcade games. It is worth noting that this distinction may have been useful in 1984, a year when around 400 video games were released worldwide–Wikipedia lists 393 titles to be exact [139]. In those days, distinguishing between three genres and leaving the result uncommented could be justified. Since then, the gaming market has grown incredibly. There have been more games and, consequently, labels that can be used to categorize them. Suffice it to say that Wolf [7] identified as many as 42 game categories. In our list of genres presented above, we limited ourselves to the 25 most popular genres, which is still an impressive number. In 2023, 13 346games were released on one platform (Steam) alone [140]. Therefore, even more questions arise: “Can the numerous game genres used by game producers and consumers be informative for a psychologist? What is behind these distinctions? Can we, as in 1984, reliably map the genres to more general characteristics of the games?”. Numerous studies have shown that MMORPGs appear to be associated with a greater risk of problematic gaming [136]. Is there a practical possibility that, based on these observations, actions that will protect users from gaming disorder will be taken to target MMORPGs as a genre? This seems unlikely for various reasons. One of the main arguments against such an action is that it is not the genre that makes the game ’dangerous’, but its features, mechanics, setting, promotion and finally the monetization model. It is worth noting that in some areas, from the very beginning, the approach from the perspective of game features rather than their genres dominates. For example, Silvern and Williamson [141], in one of the first papers on the relationship between games and aggression, write about ‘violent games’ instead of writing categories based on genres. So why not focus the search for game-located determinants of positive (e.g., cognitive development) and negative (e.g., gaming disorder) consequences of gaming on factors of psychological meaning rather than labels designed for game development and cultural studies? We postulate that the tags created by the gaming community can be an interesting alternative for genres.

4.3 A glimpse of hope

In this section, we will present the mechanism of operation of community-based tags and the possible benefits and limitations of their use as an alternative to genres. We’ll use Steam labels as an example (so-called ’Steam tags’), but many other game-selling platforms offer similar solutions that could potentially be used (e.g., GOG, Epic Games, Ubisoft Connect, or EA) [89]. Steam tags are descriptive labels assigned to games on the Steam platform. They are utilized as a way to classify games based on different aspects such as genre, gameplay features, themes, and visual properties. Steam tags are intended to improve the sale of games, with the assumption that by processing them, it becomes possible to better match the individually offered games to the needs of the recipient, which positively affects the willingness to buy [142]. Tags can be assigned by both developers and users; hence, every individual interacting with a game is able to shape its overall image through tags. The tags cover a wide range of characteristics, including genres. Some common examples include "action", "strategy" and "RPG”. However, tags can also be more specific, capturing unique aspects of a game, such as "idle", "trading card game", "co-op", "loot", " replay value", or even "addictive". There are over 420 default Steam tags (including 79 from "genre " and "61 " from "sub-genre " category), yet the list of tags is not closed, and each user can propose any tag that will be assigned to the game after approval by the moderators [142]. The number of tags assigned to a given title is unlimited, but as a result of the activity of users and publishers, the most important ones are placed first. Tags proved to be an effective way to describe games, they were introduced to Steam in 2014 and have been around ever since. It is worth noting that on this platform there is no longer a division of games by genre.

Obviously, tags (like genres) were not designed as a tool for psychological researchers; therefore, the mere fact of their existence and popularity is not a sufficient reason to depart from the traditional classification of games. However, they have several advantages over genres that may lead to interesting research results. First of all, the tags assigned to games are the result of a dynamic consensus among their developers and the gaming community. Furthermore, they are created by the community, so they reflect how players perceive and categorize games, which can give an important insight into players experiences. The tags are ordered on the game page on the basis of their popularity, which allows for easy estimation of how crucial a given characteristic is in the overall picture of a given game. At this point, however, it should be noted that the publisher can remove tags, which may make interpretation uncertain to some extent. Using steam tags may offer access to the large sample dataset, which not only covers genre but also may emphasize so-called facet characteristics of games such as gameplay, style, purpose, theme, target audience, temporal aspect, setting, and mood/affect which can be used in psychological research to distinguish individual structure characteristics that may elevate risk of gaming disorder [85,89]. Consequently, we can avoid the overpathologization of some game genres and focus on individual mechanics that may increase or decrease the risk of the potential development of behavioral addiction. Moreover, data from gaming platforms can be obtained much faster than individual and subjective categorization of games into the respective game genre. It can even be automatized. One of the main findings of this systematic review was that genre classifications are very subjective. We can also say that about game tags. However, in the first situation, a few researchers usually decide that this game will be classified as, for example, RPG. In the second situation, game developers and a large sample of the gaming community decide on tags for each game. External databases such as SteamSpy (https://steamspy.com/) provide data about how many users assigned each individual tag to an individual game. On the basis of this data, we can obtain quantitative indicators that can be further analyzed. Furthermore, using tags from worldwide accessible gaming platforms could create a way to conduct comparative research.

Even though game tags have many advantages, we still need to remember about some of the risks of using them in scientific research. First of all, game tags on each gaming platform can be diverse from each other, so relying on game tags from one specific platform may be limited. An aggregation of available data may be helpful in this case. Secondly, game tags may not present all the valuable characteristics of video games, which may have an effect on the individual’s psychological functioning. Perhaps, using more structured metadata repository like, for example, Gameopedia could be a way to cover all the important mechanics from a psychological point of view.

4.4 Gender discrepancy

This systematic review presents research over one decade and it is surprising to see that only 26% of all the participants in the studies were women. According to recent data from the Entertainment Software Association [143], 48% of players identify as women. It is a huge discrepancy between the reality presented in scientific research and the real gaming community. Such a situation may be a result of stereotypes that playing video games is primarily a male activity or the untrue belief that women play only casual games [144]. This narrow focus excludes and overlooks the experiences of women gamers, leading to a lack of representation in the research literature. On the other hand, researchers might face challenges with recruiting women participants because they may be situated outside of video gaming culture; e.g., women are engaging less with the gaming community, which has a history of being unwelcoming and discriminating towards them [144,145]. However, conducting studies on female gamers regarding their gaming preferences, risks of gaming disorder, and gaming motives is very important. Especially when recent findings show that women have very similar risk of developing gaming disorder to male gamers and that they tend to invest more time in playing video games than men [18,146].

4.5. Geographical coverage

The geographical distribution on gaming genre research seems to be heterogeneous. The most studies have been conducted in Europe (41%), which compares with different continents (Asia– 12%; Australia and Oceania– 8%; North America- 28%) generates some imbalance. Interestingly, in Asia identified only 12 studies, since this region has the largest population of players [147], so it would be expected that it will be dominant in terms of scientific research. Then, no studies have been identified in South America and Africa. Interest in game genres is not universal on all continents, therefore it can be assumed that it depends on culture variation and technological development.