Should the Monster Play Fair?: Reception of Artificial Intelligence in Alien: Isolation
by Jaroslav ŠvelchAbstract
The article addresses the gap in scholarship on reception of game AI by qualitatively analysing forum discussions about the Alien monster in the game Alien: Isolation (Creative Assembly, 2014a). The game’s Alien is a rare example of a recurrent, undefeatable AI-driven adversary, designed to represent a sublime monster. Its behaviour is governed by behaviour trees and simulated sensory perception (on the level of foreground AI), but also by the Director component, which monitors the global state of the game to maintain gameplay tension and periodically sends the creature near the player’s location (on the level of background AI). Upon the game’s release, however, very little information about its AI was disclosed. The article follows players’ efforts to theorize and evaluate the game’s AI, and their discussion about its (un)fairness. It outlines two basic approaches to enemy AI: experientialist, which focuses on how the opponent “feels,” and simulationist, which sees the monster as a simulated animal-like creature. While experientialists accept AI tricks and shortcuts, simulationists expect the Alien to be a discrete, autonomous entity, ontologically separated from its environment. The latter, for example, find it unfair that the Alien has access to information outside of its sensory perception, and seem upset by the actions of the AI Director module, which they interpret as collusion between the Alien and the game. Moreover, simulationists object to the idea that the Alien should be able to teleport or be “tethered” to the player character. The article points to an important paradox of AI in games. On the one hand, players welcome a degree of unpredictability, variation and mystery afforded by AI routines. On the other hand, many expect to be able to completely account for and understand the behaviour of computer-controlled opponents -- even if they are science fiction monsters.
Keywords: video games, artificial intelligence, AI, Alien, monster, stealth, science fiction, player studies, reception, discussion forum
Introduction
Alien: Isolation (Creative Assembly, 2014a) is a first-person stealth game that aims to recreate the tense and claustrophobic horror of the 1979 Alien film (Scott, 1979). The player assumes the role of Amanda Ripley, the daughter of the film’s protagonist Ellen Ripley. Like her mother, Amanda is confronted with the classic Alien monster, this time in the confined environment of the Sevastopol space station. The game’s Alien mimics the creature’s now-classic visual design by H. R. Giger, as well as much of its behaviour portrayed in the film. The developers advertised the monster as “an enemy that you need to avoid at all costs. An Alien that is stalking you; that’s intelligent, unpredictable, extremely dangerous.” (Creative Assembly, 2014b, pts. 2:06-2:15) Amanda cannot defeat the Alien using conventional weapons. Instead, she must learn how to anticipate, distract, and avoid the creature, using a variety of makeshift equipment. The game was disempowering by design, and purposefully elicited unpleasant emotions of fear and anxiety. To make the point, one review has described it as “a trauma machine masquerading as a video game,” adding that “only in the realm of horror could this be seen as a compliment” (Kietzmann, 2014).
In the original film, the Alien was shown in a few carefully scripted scenes, hand-crafted by the screenwriters, the director, cinematographer and the special effects team (see Turnock, 2015). Isolation, on the other hand, stars an Alien that is driven by a set of artificial intelligence (AI) techniques. Unlike many other video games enemies, who appear at fixed, pre-scripted moments in the game’s narrative, Alien’s behaviour is continuously simulated throughout the game. The AI was one of the selling points of the game, and both the developers and journalists pointed out the monster’s unpredictability (Creative Assembly, 2014b; Kietzmann, 2014). Accordingly, game AI researchers Georgios Yannakakis and Julian Togelius have included Isolation in their list of games with noteworthy application of AI (2018, p. 12). But although game AI technology is advancing, there is very little research on its reception by players. Given the game’s prominent use of AI techniques, it makes for a suitable case study to open this avenue of research.
I have investigated player discussions about Isolation’s Alien as a part of a broader research into design, representation and reception of monsters in video games, discovering a prominent and lively strand of debate concerning the “fairness” or “unfairness” of the Alien’s mechanics. Given that the Alien is a science fiction monster, equipped with superhuman speed, strength and perception, it might seem paradoxical or downright misinformed to expect it to be “fair”. Nevertheless, Isolation is still a game and -- as I will outline below -- players assume that games possess at least some degree of fairness. A closer look at player discussions reveals an important set of expectations regarding the consistency and transparency of artificial intelligence in commercial video games. It can also help us learn what kinds of behaviour players might understand as “cheating” on the part of the AI.
In this article, I will first discuss the theories of artificial intelligence and monstrosity and describe how AI works in Isolation. In the latter half of the text, I will delve into discussion forum material to investigate how players make sense of Isolation’s AI, and what they find unfair.
Illusions of Intelligence
AI has a peculiar place in digital game scholarship. On the one hand, its embrace by game technology research catapults it among the top three keywords in all Scopus-indexed game research publications (Martin, 2018). On the other hand, it is conspicuously absent from the humanistic and social scientific branches of game studies. According to the data collected by Paul Martin in 2018, the two major Scopus-indexed journals in this area of game studies -- Game Studies and Games and Culture -- had only ever published one Scopus-indexed article with AI as a keyword [1]. In a rare humanistic exploration of the topic, Fizek points out the lack of attention to non-human agency in games, including AI, and calls for “more future work… on the AI-driven automation aspect in games and virtual environments…” (2018, p. 210)
The matter is further complicated by the overall terminological and conceptual confusion in the field of artificial intelligence. Even insiders in the field have pointed out that it is “fragmented” (Eaton et al., 2016) and that the accomplishments of what is called AI are “cobbled together from a grab bag of tools and techniques” (Kaplan, 2017). In other words, AI means different things to different groups of people. In contemporary computer science, the bulk of AI research develops techniques such as neural networks for potential use in real-world applications like image recognition or language processing. In games, Yannakakis and Togelius identify three main domains for the application of AI: playing of games, generating content, and modelling of player behaviour (2018). They further divide the domain of playing games based on its role and goal. The AI can assume either a player, or a non-player role, and play either to win, or to provide an experience. A substantial part of AI research has focused on AI that aims to win in the player role, creating powerful opponents for chess, Go, or, more recently, real-time strategy games such as StarCraft II, while employing advanced machine learning techniques.
Most players, however, encounter AI that controls non-player characters (NPCs), including human and monstrous enemies [2]. As opposed to other applications, the use of AI in mainstream games is constrained by the limited processing power available to the developers, as well as by game design conventions. Video games usually get away with just an illusion of intelligence -- “for the player to believe that the NPC in some sense is intelligent even though the code controlling it is very simple” (Yannakakis & Togelius, 2018, p. 97; see also Lou, 2017). The developers’ effort is often to create the experience of an intelligent NPC rather a simulation of it. In effect, commercial games therefore tend to employ rule-based techniques such as finite state machines or behaviour trees, which do not involve learning, but execute scripted reactions to a pre-constituted set of conditions, giving the designers more authorial control than the more unpredictable learning algorithms (Yannakakis & Togelius, 2018).
Enemy AI in commercial games is, moreover, not designed to beat the player, but to entertain them by providing an adequate level of challenge. We will soon discuss Isolation players’ concerns whether the AI was fair. But what does fair mean? As Juul has pointed out, “the idea of what makes a video game ‘fair’ has changed considerably over the last few decades” based on the shifting game design conventions (Juul, 2013, p. 69). The most accepted interpretation ties fairness to skill, and boils down to the idea that that the game designer guarantees that with enough training, the player will become good enough to overcome the game’s challenges (Juul, 2013; Paul, 2018). If the AI is too powerful or too unpredictable, it carries the risk of being accused of cheating, or having “unfair advantage” over the player (Consalvo, 2009, p. 87; see Fizek, 2018) [3]. At the same time, some genres may permit more “unfairness” than others. Survival horror games like Isolation, for example, build on the idea that the odds are stacked against the player (Perron, 2018).
AI researchers have hypothesized that players conceive of AI in two basic ways. According to Combs, “[at] one extreme, people might refer to the in-game characters with which they interact as controlled by the ‘AI.’ At the other extreme, people might also speak of the invisible hand that runs the game world as controlled by the ‘AI’.” (Combs, 2004, p. 117) Treanor and colleagues similarly distinguish between foreground AI and background AI (2015). Foreground AI is associated with individual characters that the player notices and can reason about, for example NPCs or enemies. Background AI, on the other hand, consists of processes that ensure desirable gaming experience, but should not be immediately noticed by players, such as traffic behaviour in the Grand Theft Auto series that ensures that cars don’t get in your way [4]. Another example is the so-called AI Director in Left 4 Dead, a cooperative shooter game (Valve Corporation, 2008). The Director evaluates the situation of the players and maintains dramatic tension by spawning zombies for them to fight. In this case, the AI assumes the role of a referee, or a dungeon master/game master in tabletop roleplaying games.
In practice, the distinction between foreground and background AI is never clear-cut, but the typology is a useful heuristic to study the phenomenology of the game AI experience. As we will see in the Isolation case, the idea of a foreground AI is more familiar than that of the background AI. Perhaps because of the popular association between AI and robots (Yannakakis & Togelius, 2018), players might imagine that each NPC is governed by an autonomous AI routine.
The Elusive Monster
The reception of the game’s AI depends on the representation of the entity that it drives. An AI driving a human would likely be judged in relation to human intelligence. But in case of Isolation, the supposed intelligence belongs to the deadly monster from the Alien universe. In my previous work, I have suggested that the portrayal of monsters in games is marked by the tension between the sublime and contained modes of monstrosity (Švelch, 2018). The sublime mode relies on the fact that the monster is always partially unknown or cognitively challenging -- it defies being an object of our actions and rational reasoning. This approach has been traditionally deployed in myth, legends, horror fiction (notably in Lovecraftian horror) and horror film, and has been extensively theorized in philosophy (Asma, 2012; Beal, 2002; Kearney, 2002). On the other hand, contained monstrosity is one that is described, catalogued and tamed. Embodying the urge to control the contingencies of the unknown world, it can be found in medieval bestiaries or in the “monster manuals” of role-playing games like Dungeons & Dragons (Hassig, 1995; Mittman, 2006; Stang & Trammell, 2019). Video games tend to present monstrosity that is contained and demystified by translating the monsters’ features into rules and statistics. To be presentable in the simulated worlds of games, the monster’s 3D model and its range of interactions with the world need to be specified and encoded in software. The monsters become objects of player action; according to Bernard Perron, even in horror video games, “on the whole, the monster is to be faced” (Perron, 2018, p. 346).
The Alien franchise oscillates between these two poles. The original 1979 film -- which Isolation seeks to mimic -- tends towards the sublime. Stephen Mulhall sees the Alien as a force that destabilizes the anthropocentric worldview. He approaches the creature from the point of view of Ash, the ship android in the 1979 film. For Ash, the Alien “symbolizes the true significance of the cosmic principle of life; it signifies the essential insignificance of human morality and culture… -- the fact that we are not at the centre of the universe and its concerns” (Mulhall, 2016, pp. 17-18). According to Barbara Creed’s influential psychoanalytical reading, the Alien embodies the figure of an archaic monster-mother, invoking “the mysterious black hole that signifies female genitalia which threatens to give birth to equally horrific offspring as well as threatening to incorporate everything in its path” (Creed, 1993, p. 27). She points out that the image of the monster is akin to Medusa -- making the viewer look away and avoid its gaze. Brendan Keogh and Darshana Jayemanne have employed Creed’s framework to analyze Isolation, astutely observing that “[the] association of the creature in Alien with the Medusa myth is echoed… with the player’s knowledge that once the alien is spotted, it can often be too late to do anything other than wait for death and reload the game” (Keogh & Jayemanne, 2018, p. 3).
At the same time, a large part of the Alien franchise follows the (mostly doomed) efforts to contain the Alien. As Ash puts it himself, the Alien is a “perfect organism,” worthy of dissection, study and admiration. The invitation was taken up by fan communities as well as the writers of some peripheral parts of the Alien transmedia franchise, who tended to approach the Alien more encyclopedically, stating, for example, its Latin binomial species names (Internecivus Raptus) (Hutchinson, 2014). Even Ridley Scott’s 2017 film Alien: Covenant followed this logic, offering a detailed explanation of the origins of the Alien species. Some critics have, however, criticized the attempt to fill in the gaps for “cheapening” the original sense of “ancient” and “creeping” horror of unknown origins (Kamen, 2017).
To sum up, there are two basic ways of conceptualizing the Alien as a monster. First, it can be a sublime Alien, an exemplar of cosmic chaos that cannot and should not be observed and that is completely indifferent to human concerns. This version of the Alien is difficult to separate from its filmic representation. As is the case with many horror films, the Alien spends a long time out of sight, and appears in a few carefully staged and scripted scenes that preserve its mystery. The other way of looking at the Alien is zoological. (In this case, I opt to use that term rather than contained because it is never fully contained.) The zoological Alien is a creature that is nightmarishly dangerous, but still an animal species with a fixed set of features. As such, we can imagine it separately from the individual film texts and independently from the hand of the director -- as a prop in any potential media product.
Engineering the Alien
In the promotional clips and articles, Isolation developers clearly state their preference for the 1979 sublime Alien. As the team’s creative lead Al Hope put it: “When audiences went to see Alien in theaters back in 1979, they were introduced to a new creature, that was terrifying, unstoppable, and totally unique. We wanted to make a game [so] that we capture the original experience. We wanted to make the Alien scary again” (Creative Assembly, 2014b, pts. 0:33-0:55). This nostalgia-tinged rhetoric of “going back to the roots” might have been brought about by the critical and commercial failure of the previous Alien game Aliens: Colonial Marines (Gearbox Software, 2013), which was an action-packed shooter game. Creative Assembly, the British studio in charge of Isolation, was in a good position to offer a fresh start. Not only were they new to the Alien franchise. They had not even published a first-person shooter -- or any first-person game -- before.
Isolation’s gameplay consists primarily of stealth game segments, in which the goal of the character is to traverse space and fulfil tasks while avoiding detection by computer-controlled opponents (Horswill, 2014; Tremblay, Torres, & Verbrugge, 2014). In Isolation, these opponents include a handful of androids (called “Working Joes”) and humans patrolling specific locations. However, diverging from most other stealth games, Isolation features a frequently recurring major adversary -- an undefeatable Alien who follows the player throughout the game. Rather than using jump scares typical of horror games, the monster was designed to be “intelligent” and “unpredictable” [5]. In the words of the designer Gary Napper, “it doesn’t follow any prescribed path or set of behaviours… it’s just reacting to the player’s presence and the choices that the player makes.”(Creative Assembly, 2014b, pts. 1:12-1:20)
Although information about the inner workings of the game’s AI was not public at the time of the game’s release in October 2014, the studio’s Lead AI Programmer Andy Bray disclosed the principles behind the game’s AI in a July 2016 industry talk (Bray, 2016). Its content was popularized a month later in a YouTube video and blog post by Tommy Thompson, a UK-based blogger, developer and AI researcher (Thompson, 2017).
The Alien is controlled by two behaviour management modules. The first one pertains to the Alien creature itself, and can be classified as a foreground AI. Primarily, the Alien stalks the station searching for prey, and reacts to the information it can perceive by its simulated sensors, attacking Ripley or other humans when it notices them [6]. The Alien also seems to be able to learn from player behaviour, for example peeking into lockers if the player has repeatedly used these as hiding spots. However, rather than a true learning AI, the game employs a clever trick. The Alien’s behaviour is governed by a decision tree, which contains hierarchically organized behaviours that trigger under certain conditions, such as running to search a location when detecting noise. Some of these behaviours are inactive at the beginning of the game, but unlock as the game progresses, creating the illusion that the Alien is learning.
The second module is the Director, which monitors the player’s situation and maintains gameplay tension, and therefore fits into the category of background AI. It helps achieve the experience the developers have called “psychopathic serendipity,” meaning that the monster appears whenever the player might have thought it was gone for good. After an unsuccessful search or after being chased away, the Alien disappears into ceiling vents (also called backstage by the developers). The program monitors player situation using a metric called the menace gauge. If it has been low for a set duration, the Director will give the Alien information about the rough whereabouts of the player, leading to an encounter. Thompson summed it up with this metaphor: “The [Alien] in Alien: Isolation has two brains: one that always knows where you are and gives hints to the second that controls the body” (Thompson, 2017). This is a crucial point where the background AI and foreground AI merge -- and the monster is periodically colluding with the “game.” As Keogh and Jayemanne have put it, “Isolation’s alien feels less like a ludic challenge to learn, overcome and enjoy, and more like an intentional, unpredictable and malicious glitch in the system to be avoided at all cost.” (Keogh & Jayemanne, 2018, p. 10)
The disclosure of Isolation’s AI gives us insight into the ways in which illusions of intelligence and experiences of monstrosity are created in games. We can see that the designers’ efforts only partially resemble scientists trying to simulate a (non-existent) zoological species. They are more akin to film industry’s special effects personnel, manipulating and tricking the player. It is no coincidence that the terminology used for these manipulations (backstage, director) is borrowed from film. Importantly, players did not initially have access to the information regarding the inner workings of the AI modules. While the presence of a refined foreground AI could be derived from the title’s promotional discourse about the game, the existence of the Director or the pseudo-learning technique could not. The behaviour of the Alien could therefore appear mysterious and surprising to first-time players.
This experience is further enhanced by the game’s recreation of the technology depicted in the 1979 film (Sloan, 2016). Ripley’s tools are low-tech, including the most important one, the motion tracker. Not only can the Alien hear its beeping, but its limited range and resolution convey the idea that the station always contains spaces that are unaccounted for.
Methodology
This article is interested in how players make sense of game AI. As such, it requires a model of how players gain knowledge of game mechanics in general. Dominic Arsenault and Bernard Perron (2009) view the process of gameplay as a feedback loop, in which the player interacts with the game and observes the outcome. The player approaches the game with existing knowledge about the title, the genre or game design conventions. This information is being continuously revised as the play progresses -- and the amount of knowledge increases -- in what Perron has called a heuristic circle (Perron, 2006). However, the player typically does not have direct access to the game code, and their image of the game (which Arsenault and Perron call game’) never fully overlaps with the game as it was programmed. As Daniel Vella argues, “[the player] is constantly drawn to confront the necessarily tentative nature, not only of her interpretation of her experience of the game into the ordered form of a cosmos, but of her direct phenomenal experience of the game object itself” (Vella, 2015). In his own work, Vella focuses specifically on the aesthetic experience of not being able to grasp the rules (or other elements) of the game -- a phenomenon he calls the ludic sublime. He gives the example of Dark Souls (FromSoftware, 2011), a game that deliberately obfuscates its rule systems and item descriptions, evoking a sense of mystery (Vella, 2015). Although Isolation is not as thoroughly confounding as Dark Souls, it did aim to re-mystify the Alien and present it as an unknown quantity, a sublime and unpredictable monster.
Although neither Arsenault and Perron, nor Vella, place emphasis on the social aspects of gameplay, reception of a game does not unfold in a vacuum. Players read tips and comments in gaming media, social media, wikis or streaming platforms; they share their own gameplay experiences and achievements as well as the questions they have about the game. In the case of Isolation, the game’s high difficulty provided a significant incentive to do so [7]. Collectively, a community of players may thus get very close to figuring out even the most obscure of game mechanics. Purposeful reverse-engineering of game mechanics is sometimes described by the term theorycrafting, defined by Christopher Paul as “the search for the optimal set of strategies with which to play.” (Paul, 2011) However, this article investigates the discourse about Isolation at a stage where the players’ understanding of the game was still limited -- to investigate their unfinished and untested, rather than finely crafted, theories.
My main research questions in this project were the following: (1) How do players theorize about and evaluate the Alien AI? (2) How are their theories and evolutions aligned to the ideas of sublime and contained monstrosity, and foreground and background AI? I chose to study these questions qualitatively, using material from online discussion forums, which allow for more structured, more precisely formulated and dialogical discussion than other platforms such as YouTube. I collected forum content published between October 7, 2014, (Isolation’s release date) and October 31, 2017 (date of data collection) from the three following sites:
- Steam [S]. Steam was, at the time of data collection, the dominant digital distribution platform for PC gaming, and its forums serve as hubs of discussion among players on PC as well as other platforms. Out of my three sites, the Steam forum had by far the most total traffic. Each game sold through the platform, including Isolation, has its own forum. To retrieve older threads, one must use the search function. I used the terms “alien behavio(u)r” and “alien AI” to retrieve 55 threads relevant to the research questions.
- AVPGalaxy [A]. Its tagline being “The Pulse of the Alien and Predator Community,” this is a community-run website that serves the fans of the Alien and Predator franchises, including the games. It contains a separate forum for Given the site’s focus, we can expect the forum users to be knowledgeable about the whole franchise. From the listing of all available discussion threads, I selected 33 that focused on the topic of Alien AI.
- RPGCodex [R]. RPGCodex is a long-running forum hosted by a community of computer role-playing games (RPG) fans. It has become well-known for its members’ criticism of what they perceive as the “dumbing down” of contemporary mainstream games. I have included RPGCodex’s one massive thread on Isolation because it was referenced on AVPGalaxy. Out of the thread’s total 62 pages, I selected 35 pages that discussed Alien AI.
From this searchable corpus, I excerpted individual posts that were most closely related to my research questions, arriving at a smaller collection of 42 posts, which I coded thematically (Ayres, 2008) and analyzed following the principles of analytic induction (Preissle, 2008). To comply with the guidelines of the project’s funding body as well as the ethical recommendations of the Association of Internet Researchers (Markham & Buchanan, 2012), I pseudonymized all participants and paraphrased the source material.
At certain points, I will contrast the players’ theories with the developers’ description of the game systems. I do not, however, have access to the game’s source code, and I do not imply that the developers are always truthful. I am more interested in what and how is being discussed than its factual merits. In addition to discussions, I also examined player mods or modifications, which were also referenced on the forums. Modifications provide a useful lens to study video game reception, because they tend to address players’ preferences and concerns (Postigo, 2016). Through modification, users can also get a glimpse of how the game works.
Reception of AI in Isolation
As an almost ever-present antagonist, Isolation’s Alien is under significant scrutiny from the community. In general, the players’ discourse about the Alien AI unfolds on four levels -- (1) they trade their stories about it; (2) they share tips and strategies regarding how to play the game successfully; (3) they try to understand and theorize how the AI works [8]; and (4) they evaluate the AI and their experience of it. I will focus primarily on the latter two categories of discourse, although they are often mixed with the former two.
Although I am more concerned with the content rather than the form of the contributions, it is worth noting that the Alien usually occupies the subject position of sentences (the Alien learns/attacks/teleports), unless there is a specific reference to the game software, AI routines or developers, which can also function as subjects (the game/system/AI does X). In some posts, the expressions “Alien” and “AI” are used interchangeably, suggesting a strong level of identification of the creature and the game’s AI routines.
Based on their approach to gameplay, players commenting on the Alien can be roughly divided into two camps, which I will call experientialists and simulationists [9]. The former see the game primarily as an experience, while the latter see it as a simulated system. Experientialists are primarily interested in how the Alien “feels,” their point of reference often being the 1979 film. In their view, a monster should remain sublime; and too much theorizing ruins the game and breaks the suspension of disbelief. An experientialist commenter suggested that humans are not supposed to understand the Alien, because it is a creature with its own “alien logic” and “quirks” [R]. In Kirsi Kallio and her colleagues’ typology of gamer mentalities, experientialists would fall somewhere between the “entertaining” and “immersing” profiles (Kallio, Mäyrä, & Kaipainen, 2011). They consider games just one of their possible entertainment options, connect and compare them to film, and immerse themselves in the game world rather than optimize their play strategies.
Simulationists, on the other hand, observe and scrutinize the game’s mechanics, mostly in order to get better at the game. To do so, they expect the mechanics to be consistent and transparent. In the abovementioned typology, they would correspond to the “having fun” profile of players for whom “speed, progress, flow, skillfulness, and other such characteristics of a game are more important than the story or… characters” (Kallio et al., 2011, p. 344). As such, they are more likely to conceive of the Alien as a simulated zoological entity.
Overall, experientialist comments tend to be in the minority, and often assume a defensive position. But that does not mean that this approach to the game is less frequent. In the case of Isolation, simulationists may simply have more matters to discuss and criticize, and more puzzles to crack. These two groups are not clear-cut, and should be understood as sets of discursive subject positions rather than individuals. Neither are they completely at odds with each other [10]. Both groups tend to enjoy the Alien’s unscripted nature, and the varied gameplay it generates. Simulationists appreciate that the monster can “surprise” them [S], while experientialists praise it for feeling like a “real creature” [S]. They differ, however, in evaluating whether the Alien is fair.
Players writing from the simulationist perspective complained that the game did not feel “fair” and that the Alien “cheated.” The experientialist retort was that expecting fairness goes against the source material -- after all, monsters are not supposed to be fair. One participant argued that Isolation should make you feel like Ellen Ripley, who was unlikely to feel “comfortable” or “fairly treated” during her ordeals [S]. There were three main aspects of the Alien’s behavior that simulationists found unfair: (1) the Alien’s “psychic” knowledge of the player’s location, (2) the Alien’s presumed teleportation, (3) the fact that the Alien is “tethered” to the player character. These will be discussed one by one in the following sections. Interestingly, the pseudo-learning component of the Alien’s AI was never criticized.
Telepathic Alien
Soon after the game’s release, players started noticing that the Alien seemed to know their whereabouts. It was accused of being “psychic” [S] or having a Ripley “homing beacon.” [S] This was considered an unfair advantage over the player. Although the Alien is clearly portrayed as faster, stronger and larger than Ripley, some players believed that the chances of winning should at least be balanced by the two opponents having the same amount of information. Instead, they felt like they were playing hide and seek against an opponent who knew their location [S].
From the simulationist point of view, the Alien should respond exclusively to “correct stimulus.” [S] This highlights the idea of a “fair” computer-controlled enemy as an autonomous entity that is dropped into the game world and navigates it with the help of its own algorithms, which are separated from the rest of the game. Following this logic, using any information that is outside of the enemy’s audiovisual perception would be considered cheating. However, this opinion, while widespread, was not shared by all users. Its critics pointed to the clues in the Alien franchise that suggest the Alien is telepathic or can track pheromones [11].
We now know that this “unfair” advantage was brought about by the macro-level Director AI, which points the Alien towards the player. Although this information had not been revealed by the time these discussions took place, a handful of players got close to figuring out the puzzle in a few isolated posts. One of the users identified the dual nature of the monster, calling it “game/alien” and suggesting that it knew where Ripley was, and patrolled that area until it found her [S]. Already about a week after the release, another Steam user, likely a programmer by training, insisted there was “something” that kept sending the Alien into the general vicinity of the player “from time to time,” because otherwise the monster would confront the player much less often [S] [12]. However, this finding did not gain wide recognition.
Teleportation Theory
Puzzled by the feeling that the Alien was “always in the proximity” of their hiding spots [A], players discussed theories of the Alien’s movement within the game world; building on concepts, terms, and metaphors known from other games, such as teleportation. Proponents of the teleportation theory argued that the Alien was teleporting, warping, or spawning near the player. This idea provoked lengthy and passionate discussions and at least two dedicated threads. Its supporters posted videos to back up their claims, pointing out that the motion tracker suddenly lost track of the monster, who had climbed into a ceiling vent and surprised Ripley from behind. Others argued that the ceiling ducts are “imaginary” and function like teleports [S]. The detractors, on the other hand, rightly pointed out that the motion tracker has a very limited reach, and brought up Alien’s superhuman speed [13]. The matter was eventually clarified by the developers, who have claimed that the Alien is only teleported twice in the course of the game, in order to appear in cutscenes (Thompson, 2017). When the Alien is in the ducts, it follows a specific set of “backstage” behaviours, but, reportedly, its movement though the space station is constantly simulated.
Although the teleportation theory has been refuted, it shows that players are invested in the idea of the monster’s permanence and contiguous movement. Teleportation was not only considered unfair, but also “cheap” [S] -- it would have been a simple, unsophisticated trick. Moreover, unlike in the case of telepathy, the Alien franchise did not offer fictional justification for the Alien’s teleportation.
Tether Theory
The tether hypothesis assumed that the Alien is connected to Ripley by an imaginary tether, which prevents it from venturing too far from her and ensures that it is constantly “stalking” [S] her [14]. On several occasions, it was likened to rubber banding, a racing game mechanic that slows down computer-controlled racers when they are in the lead and speeds them up when they are behind, so that the race remains competitive. The term tether itself might have been inspired by the lukewarm review of Isolation published by the gaming site Polygon, which argued that the Alien “seems attached to you by some kind of invisible tether” (Gies, 2014). Like the reviewer, some players found tethering frustrating, as it created too many tiresome and tedious encounters with the monster. Simulationists found it at odds with the promise of an independent, intelligent Alien. However, some experientialists defended this design choice, attributing it to the “strange symbiosis” [S] between Ripley and the Alien, as it is portrayed in the 1979 film.
Unlike teleportation, the tether theory has a basis in the game’s software. The modding community discovered a set of variables called “exclusion radius,” which affect the radius within which the Alien will wander around the player (Nathan, [TW]Tingbemine, & MattFiler, 2018). These variables were tweaked by subsequent mods.
Modding the Alien
Mods for Isolation aimed to address some of the players’ concerns. However, Isolation did not have built-in mod tools, and it took the modders months to map the game’s data structure. Interestingly, the first mod requested by the player community asked for the removal of the Alien, because some players found it too difficult or annoying. As this was a relatively crude change that did not require intimate knowledge of game files, the Alien was successfully removed from the game already on November 1, 2014, making it the first mod to appear in my material. The mod was met with predictable derision from players who saw the Alien as integral to the game and likened the Alien-less Isolation to Gone Home (The Fullbright Company, 2013), a “walking simulator” game criticized by some members of core gamer communities for offering little challenge (see Pavlounis, 2016). Others, however, enjoyed it as a new and refreshing way of playing the game. Fans of the film, especially, enjoyed exploring the game as an “Alien museum” and marvelling at its set design [R]. Instead of reliving Ripley’s experience, they could enjoy an experience similar to walking around on the 1979 Alien film set.
The first comprehensive Isolation modding guide was uploaded to Steam in April 2015. Around that time, one of its co-authors released the Unpredictable Alien mod (somethingelse666, 2015), by far the most downloaded Isolation mod [15]. The mod was supposed to “enhance” the game by doubling down on the designers’ promise to create an unpredictable monster. It extended the creature’s stalking radius and loosened the tether. The mod’s reception was mixed. From the simulationist point of view, it was praised it for making Alien’s behaviour more “natural.” To some experientialists, it granted a more sublime experience of not knowing “the world’s limitations” [R]. However, it seemed to break some of the other game mechanics and gameplay strategies, leading one user to conclude that the Unpredictable Alien mod is “too unpredictable.” [R] Moreover, it was difficult to test whether the mod worked properly. The Alien behaviour turned out to be interlocked with many different aspects of the game, some of which could not be modded because they were either hard-coded in the program, or closely dependent on the maps of the game world. The idea that the Alien can be, figuratively “set free” to become an autonomous creature proved false.
Conclusions
The material has revealed that the reception of the Alien AI was quite diverse. This diversity reflects several factors: the unpredictable nature of the AI, which can lead to different outcomes for different players; the multiplicity of different player mentalities and strategies; and also the divergent portrayals of the Alien in the transmedia franchise, which range from sublime to zoological, and can be used to argue for or against various features of Isolation’s Alien (see Table 1).
Players speaking from the experientialist position tend to be tolerant of the tricks and shortcuts that make the AI work, and comfortable with the fact that a game is full of illusions and “smoke and mirrors” [A]. In a way, experientialist players conceptualize the game similarly to a film, respecting the right of the game designers to manipulate them in order to provide a desirable experience. They are ready to accept the presence of a director (or an AI Director), and acknowledge the fact that monsters and NPCs are puppets controlled by this director. When the monster seems to be “psychic” or “tethered,” experientialists can explain it by appealing to the creature’s sublime qualities. This does not necessarily mean that they enjoy the game more. In fact, given that they are less focused on mastering the game’s mechanics, they might be likely to abandon the title because of its numerous encounters with the monster -- which can be difficult if one has not come up with efficient gameplay strategies.
Players who align with the simulationist viewpoint expect a more consistent and straightforward set of mechanics, or, as Vella has put it, a “perfectly-ordered, rule-bound cosmos” (Vella, 2015). This tendency is most apparent in their critique of the Alien’s presumed telepathy, teleportation, and tethering. They imagine it as an embodied creature that should be constantly observable in the simulated world of Sevastopol. Any dislocation that contravenes Newtonian mechanics (teleportation or tethering) is considered cheating. So is access to general, perception-independent information about the state of the game world. Simulationists expect the Alien to be a discrete, autonomous entity, ontologically separated from its environment. Rather than a sublime monster, they conceive of the Alien as an animal-like robot that runs on its own AI routine, which is independent of the game’s main software. But despite their insistence on the monster’s independence, they evaluate its sensory perception from an anthropocentric point of view, expecting that the creature’s perception of the world roughly corresponds to theirs [16].
Table 1. Overview of the experientialist and simulationist positions. Experientialists do not explicitly discuss the possibility of teleportation in my material, but since they do not defend it, we may expect them to consider it unfair.
Out of the AI techniques used in Isolation, the quasi-learning component was received mostly positively and without much controversy, as it is consistent with the transmedia franchise, and does not compromise the autonomy of the creature. However, the merging of background and foreground AI created considerable confusion. The idea of background AI is not widely familiar, and the blending of foreground and background AI seems counter-intuitive. While some players accepted it as a form of monstrous intelligence, others criticized it as an intelligence that was quasi-human or animal but cheating. The stance towards background AI might, of course, vary depending on genre or individual titles. In strategy or sports games, players might be more open to the idea that they are playing “against the computer,” and that this computer is programmed to make “intelligent” moves on behalf of the opposing army or team. However, returning to Yannakakis’ and Togelius’ typology (2018), Isolation makes it somewhat unclear whether the AI occupies a player or a non-player position. We seem to be playing against the Alien and against the computer (or the Director) at the same time, which some players may consider unfair. This leads us to the issue of transparency. The perception of unfairness may stem from a breach of an unwritten contract between the players and the developers. Players may feel tricked or frustrated when developers withhold crucial information about AI routines. In Left 4 Dead, after all, the inclusion of background AI “Director” was explicit, and that might have prevented confusion. On the other hand, revealing such techniques may compromise the game’s mystery. Left 4 Dead offered a very different experience from Isolation; one in which mystery and the sublime were sidelined.
As they have been derived from empirical material covering one specific game, the experientialist and simulationist viewpoints are not meant to replace existing player typologies but could prove useful when studying reception of AI in other titles. Most mainstream video games today aspire to provide both simulations and experiences, and the requirements for the two can often be at odds. On the one hand, a simulation is presumably objective, running the game world independently of the players. Ensuring a desirable experience, on the other hand, puts the player subject in the centre, anticipating or monitoring their situation, and adjusting to it. Often, the consistency of simulation must be compromised in order to produce a desired experience, disappointing simulationist players. Isolation strove to provide the experience of “psychopathic serendipity” and sublime monstrosity, at the expense of the simulation of a zoological alien. As video games diminish the sublime qualities of monsters by modelling them in simulated game worlds (Švelch, 2013), the obfuscation of rules or the collusion between the monster and other game systems are among the few techniques that make monsters sublime, and that was also the path taken by Isolation.
To this day, several aspects of Isolation’s design remain original and unique. The game’s reception was positive but divided, with some journalists finding the Alien encounters repetitive and annoying (Gies, 2014; Marsh, 2014). Due to disappointing sales, Sega did not commission a sequel (Handrahan, 2015). It is difficult to assess whether the AI played any role in its somewhat muted reception, but the relative lack of commercial success might have made the idea of a recurring and powerful AI-controlled opponent seem less attractive to publishers. As it is, the game highlights an important paradox surrounding AI in games. On the one hand, players welcome a degree of unpredictability, variation and mystery. On the other hand, many expect to be able to completely account for and understand the behaviour of enemies and non-player characters -- even if they are science fiction monsters.
Acknowledgments
The article has benefited from generous advice and feedback provided by Kristine Jørgensen, Kristian A. Bjørkelo, Daniel Vella, Jan Švelch, and two anonymous reviewers. The research for the article was supported by the Research Council of Norway through the Games and Transgressive Aesthetics project undertaken at the University of Bergen. An early version of this work was presented at the Games, Values and AI workshop organized by the Leverhulme Centre for the Future of Intelligence at the University of Cambridge. I would like to thank the participants and organizers for their tips and feedback.
Endnotes
[1] The unpublished dataset for (Martin, 2018) was kindly provided by Paul Martin. The one identified article was published in Game Studies and discusses AI only tangentially (Hutchison, 2008). The dataset does not include two articles with the keyword AI from the Simulation & Gaming journal, which publishes both game technology and humanistic/social-scientific research (Yilmaz, Ören, & Aghaee, 2006; McKenzie et al., 2008). This is probably due to an error in the Scopus database’s search function. As of May 15, 2020, these two articles are present in the database but do not appear as search results for the given keyword(s). Both articles, however, feature applied research rather than humanistic or social scientific exploration of the topic.
[2] In the vernacular of game developers, the term “game AI” tends to refer to any program code that controls NPCs or enemies, no matter if they are supposed to be intelligent (Yannakakis & Togelius, 2018).
[3] Although game scholarship has mostly focused on cheating by players (Consalvo, 2009), such unfair advantage can also be perceived in the case of computer-controlled opponents.
[4] Some conceptualizations of game AI, for instance by Mateas, require “intelligence” to be recognized as such by the player. In that case, background AI would not even be considered AI (Mateas, 2003).
[5] Although many contemporary games contain numerous types and exemplars of monsters, Isolation can be considered an inheritor to an overlooked lineage of games in which the main character faces one foe within a labyrinth. These include Hunt the Wumpus (Yob, 1973) and 3D Monster Maze (Evans & Greye, 1982). See also (Perron, 2018). Another more contemporary example of a one-monster game is Miasmata (IonFX, 2012).
[6] Interestingly, the sensors are reportedly placed on the back as well as the front of its head to prevent the players from sneaking around the Alien’s back. In the Alien franchise, a hypothesis exists that the Alien detects the electrical fields around people’s bodies. This hypothesis originates in the novelization of the original film, and is attributed to the character of Ellen Ripley (Foster, O’Bannon, & Shusett, 2014).
[7] In January 2015, three months after the game’s release, only 15.1% players who owned the title on Playstation 3 or PlayStation 4 finished the game (PSMania, 2015).
[8] I call them theories, but the participants do not necessarily use the word.
[9] This empirical observation somewhat aligns with Torben Grodal’s theoretical dichotomy of “game as an experiential route” and “game as a map and as a system.” (Grodal, 2003)
[10] A hypothetical game in which the player, for example, fights conventional killer robots (portrayed as contained, rather than sublime, monsters) with consistent and transparent AI would be much less ambitious than Isolation but would likely alienate neither experientialists nor simulationists.
[11] The mechanics of Aliens sensing pheromones featured heavily in the 2010 game Aliens vs. Predator (Rebellion Developments, 2010).
[12] The user’s profile is linked to their blog about programming.
[13] According to one of the loading screen tips, “the motion tracker is only accurate in the direction you are facing.” (Creative Assembly, 2014a)
[14] A variation on tethering is the habitat theory, proposed in one of the fan-produced guides published on Steam (Valeryn, 2014).
[15] As of May 15, 2020, it has registered around 20,000 downloads on the Mod DB database. The second most downloaded mod, Alien Ignores Amanda (DrReiCow, 2016), has about 5,000 downloads.
[16] Similarly, some of the research on humans’ attitudes toward robots has shown that people tend to expect robots to be humanoid (Nomura, Kanda, Suzuki, & Kato, 2005).
References
Arsenault, D., & Perron, B. (2009). In the Frame of the Magic Cycle: The Circle(s) of Gameplay. In B. Perron & M. J. P. Wolf (Eds.), The video game theory reader 2 (pp. 109-132). New York: Routledge.
Asma, S. T. (2012). On monsters: An unnatural history of our worst fears. Oxford: Oxford University Press.
Ayres, L. (2008). Thematic Coding and Analysis. In L. M. Given (Ed.), The Sage encyclopedia of qualitative research methods (pp. 867-868). Los Angeles: Sage Publications.
Beal, T. K. (2002). Religion and its monsters. New York: Routledge.
Bray, A. (2016, July). It’s in the Vents: The AI of Alien Isolation. Retrieved February 19, 2019, from Nucl.ai Archives website: https://archives.nucl.ai/recording/its-in-the-vents-the-ai-of-alien-isolation/
Combs, N. (2004). Artificial intelligence and the narrative experience in the virtual world. On the Horizon, 12(3), 117-122. https://doi.org/10.1108/10748120410555368
Consalvo, M. (2009). Cheating: Gaining Advantage in Videogames. Cambridge, MA: MIT Press.
Creative Assembly. (2014b). Alien: Isolation Developer Diary: “Creating The Alien.” Retrieved from https://www.youtube.com/watch?v=6W6OpHB2D4Q
Creed, B. (1993). The monstrous-feminine: Film, feminism, psychoanalysis. London: Routledge.
DrReiCow. (2016). Alien Ignores Amanda. Retrieved December 4, 2017, from Mod DB website: https://www.moddb.com/mods/alien-ignores-amanda
Eaton, E., Stone, P., Walsh, T., Wooldridge, M., Dietterich, T., Gini, M., … Russell, S. (2016). Who speaks for AI? AI Matters, 2(2), 4-14. https://doi.org/10.1145/2847557.2847559
Fizek, S. (2018). Automated State of Play. Rethinking Anthropocentric Rules of the Game. In R. Reichert, M. Fuchs, P. Abend, A. Richterich, & K. Wenz (Eds.), Rethinking Al: Natural networks, biometrics and the new artificial intelligence. Bielefeld: transcript.
Foster, A. D., O’Bannon, D., & Shusett, R. (2014). Alien: The official movie novelization. London: Titan.
Gies, A. (2014, October 3). Alien Isolation Review: Crew Expendable. Retrieved from Polygon website: https://www.polygon.com/2014/10/3/6142209/alien-isolation-review-xbox-one-PS4
Grodal, T. (2003). Stories for Eye, Ear, and Muscles: Video Games, Media, and Embodied Experiences. In M. J. P. Wolf & B. Perron (Eds.), The video game theory reader (pp. 129-155). New York: Routledge.
Handrahan, M. (2015, May 11). Sega laments “weak” game sales even as hard numbers rise. Retrieved from Gamesindustry.biz website: https://www.gamesindustry.biz/articles/2015-05-11-sega-laments-weak-game-sales-even-as-hard-numbers-rise
Hassig, D. (1995). Medieval Bestiaries: Text, Image, Ideology. Cambridge: Cambridge University Press.
Horswill, I. (2014). Game design for classical AI. Tenth Artificial Intelligence and Interactive Digital Entertainment Conference.
Hutchinson, L. (2014, August 2). The throwaway line in Aliens that spawned decades of confusion. Retrieved from Ars Technica website: https://arstechnica.com/the-multiverse/2014/08/the-throwaway-line-in-aliens-that-spawned-decades-of-confusion/
Hutchison, A. (2008). Making the Water Move: Techno-Historic Limits in the Game Aesthetics of Myst and Doom. Game Studies, 8(1). Retrieved from http://gamestudies.org/0801/articles/hutch
Juul, J. (2013). The art of failure: An essay on the pain of playing video games. Cambridge, MA: MIT Press.
Kallio, K. P., Mäyrä, F., & Kaipainen, K. (2011). At Least Nine Ways to Play: Approaching Gamer Mentalities. Games and Culture, 6(4), 327-353. https://doi.org/10.1177/1555412010391089
Kamen, M. (2017, May 7). Alien: Covenant review: a stupid, disappointing mess that cheapens the original franchise. Wired UK. Retrieved from https://www.wired.co.uk/article/alien-covenant-review
Kaplan, J. (2017, March 3). AI’s PR Problem. Retrieved September 13, 2018, from MIT Technology Review website: https://www.technologyreview.com/s/603761/ais-pr-problem/
Kearney, R. (2002). Strangers, Gods, and Monsters: Interpreting Otherness. London: Routledge.
Keogh, B., & Jayemanne, D. (2018). “Game Over, Man. Game Over”: Looking at the Alien in Film and Videogames. Arts, 7(3), 43. https://doi.org/10.3390/arts7030043
Kietzmann, L. (2014, March 10). Alien: Isolation review: Escape claws. Retrieved February 26, 2019, from Engadget website: https://www.engadget.com/2014/10/03/alien-isolation-review-escape-claws/
Lou, H. (2017, August 28). AI in Video Games: Toward a More Intelligent Game. Retrieved September 13, 2018, from Harvard University, The Graduate School of Arts and Sciences: Science in the News website: http://sitn.hms.harvard.edu/flash/2017/ai-video-games-toward-intelligent-game/
Markham, A., & Buchanan, E. (2012). Ethical Decision-Making and Internet Research, version 2.0. Retrieved from http://aoir.org/reports/ethics2.pdf
Marsh, C. (2014, October 16). Alien: Isolation is a stunningly realistic locker simulator. Retrieved February 26, 2019, from AV/ Games website: https://games.avclub.com/alien-isolation-is-a-stunningly-realistic-locker-simul-1798273066
Martin, P. (2018). The Intellectual Structure of Game Research. Game Studies, 18(1). Retrieved from http://gamestudies.org/1801/articles/paul_martin
Mateas, M. (2003, November 4). Expressive AI: Games and Artificial Intelligence. Presented at the Level Up: Digital Games Research Conference, Utrecht. Retrieved from http://users.soe.ucsc.edu/~michaelm/publications/mateas-digra2003.pdf
McKenzie, F. D., Petty, M. D., Kruszewski, P. A., Gaskins, R. C., Nguyen, Q.-A. H., Seevinck, J., & Weisel, E. W. (2008). Integrating crowd-behavior modeling into military simulation using game technology. Simulation & Gaming, 39(1), 10-38. https://doi.org/10.1177/1046878107308092
Mittman, A. S. (2006). Maps and monsters in medieval England. New York: Routledge.
Mulhall, S. (2016). On film (Third edition). London: Routledge.
Nathan, [TW]Tingbemine, & MattFiler. (2018, September 10). Modding Alien: Isolation. Retrieved February 12, 2019, from Steam website: https://steamcommunity.com/sharedfiles/filedetails/?id=423489181
Nomura, T., Kanda, T., Suzuki, T., & Kato, K. (2005). People’s assumptions about robots: Investigation of their relationships with attitudes and emotions toward robots. ROMAN 2005. IEEE International Workshop on Robot and Human Interactive Communication, 2005, 125-130. https://doi.org/10.1109/ROMAN.2005.1513768
Paul, C. A. (2011). Optimizing Play: How Theorycraft Changes Gameplay and Design. Game Studies, 11(2). Retrieved from http://gamestudies.org/1102/articles/paul
Paul, C. A. (2018). The toxic meritocracy of video games: Why gaming culture is the worst. Minneapolis: University of Minnesota Press.
Pavlounis, D. (2016). Straightening Up the Archive: Queer Historiography, Queer Play, and the Archival Politics of Gone Home. Television & New Media, 17(7), 579-594. https://doi.org/10.1177/1527476416631627
Perron, B. (2006). The Heuristic Circle of Gameplay: The Case of Survival Horror. In M. Santorineos (Ed.), Gaming Realities: A Challenge of Digital Culture. Retrieved from https://www.ludov.ca/documents/publications/Perron%20-%20Heuristic%20Circle%20of%20Gameplay%20-%20Mediaterra%202006.pdf
Perron, B. (2018). The world of scary video games: A study in videoludic horror. New York: Bloomsbury Academic.
Postigo, H. (2016). Modification. In R. Guins & H. Lowood (Eds.), Debugging game history: A critical lexicon (pp. 325-334). Cambridge, MA: MIT Press.
Preissle, J. (2008). Analytic Induction. In L. M. Given (Ed.), The Sage encyclopedia of qualitative research methods (pp. 15-16). Los Angeles, CA: Sage Publications.
PSMania. (2015, January 26). You’ll be surprised how few gamers actually finished Alien: Isolation. Retrieved February 19, 2019, from PlayStation Mania -- Games, news, reviews, interviews and trophies website: https://www.psmania.net/2015/01/youll-be-surprised-how-few-gamers-actually-finished-alien-isolation-was-the-game-too-well-made
Scott, R. (1979). Alien. 20th Century Fox.
Sloan, R. (2016). Homesick for the unheimlich: Back to the uncanny future in Alien: Isolation. Journal of Gaming & Virtual Worlds, 8(3), 211-230. https://doi.org/10.1386/jgvw.8.3.211_1
somethingelse666. (2015). Unpredictable Alien mod for Alien: Isolation. Retrieved December 4, 2017, from Mod DB website: http://www.moddb.com/mods/unpredictable-alien-mod
Stang, S., & Trammell, A. (2019). The Ludic Bestiary: Misogynistic Tropes of Female Monstrosity in Dungeons & Dragons. Games and Culture, 155541201985005. https://doi.org/10.1177/1555412019850059
Švelch, J. (2013). Monsters by the numbers: Controlling monstrosity in video games. In M. Levina & D.-M. T. Bui (Eds.), Monster culture in the 21st century: A reader (pp. 193-208). New York: Bloomsbury Academic.
Švelch, J. (2018). Encoding monsters: “Ontology of the enemy” and containment of the unknown in role-playing games. Presented at the Philosophy of Computer Games 2018, Copenhagen. Retrieved from http://gameconference.itu.dk/papers/09%20-%20svelch%20-%20encoding%20monsters.pdf
Thompson, T. (2017, April 24). The Perfect Organism. Retrieved September 21, 2017, from Becoming Human website: https://becominghuman.ai/the-perfect-organism-d350c05d8960
Treanor, M., Zook, A., Eladhari, M. P., Togelius, J., Smith, G., Cook, M., … Smith, A. (2015). AI-based game design patterns. In Proceedings of the 10th International Conference on the Foundations of Digital Games 2015 (FDG 2015). Retrieved from https://strathprints.strath.ac.uk/57219/
Tremblay, J., Torres, P. A., & Verbrugge, C. (2014). Measuring risk in stealth games. Proceedings of the 10th International Conference on the Foundations of Digital Games 2014 (FDG 2014). Retrieved from http://www.fdg2014.org/papers/fdg2014_paper_33.pdf
Turnock, J. A. (2015). Plastic reality: Special effects, technology, and the emergence of 1970s blockbuster aesthetics. New York: Columbia University Press.
Valeryn. (2014, October 10). The Alien’s AI (VALERYN’s guide in English) (h0rnyfavn, Trans.). Retrieved February 12, 2019, from Steam website: https://steamcommunity.com/sharedfiles/filedetails/?id=323874059
Vella, D. (2015). No Mastery Without Mystery: Dark Souls and the Ludic Sublime. Game Studies, 15(1). Retrieved from http://gamestudies.org/1501/articles/vella
Yannakakis, G. N., & Togelius, J. (2018). Artificial Intelligence and Games. Cham: Springer International Publishing.
Yilmaz, L., Ören, T., & Aghaee, N.-G. (2006). Intelligent agents, simulation, and gaming. Simulation & Gaming, 37(3), 339-349. https://doi.org/10.1177/1046878106289089
Ludography
Creative Assembly. (2014a). Alien: Isolation [PC]. Sega.
Evans, M., & Greye, J. K. (1982). 3D Monster Maze [ZX81]. J. K. Greye Software.
FromSoftware. (2011). Dark Souls [PS3]. Namco Bandai.
The Fullbright Company. (2013). Gone Home [PC]. The Fullbright Company.
Gearbox Software. (2013). Aliens: Colonial Marines [PC]. Sega.
IonFX. (2012). Miasmata [PC]. Devolver Digital.
Rebellion Developments. (2010). Aliens vs. Predator [PC]. Sega.
Valve Corporation. (2008). Left 4 Dead [PC]. Valve Corporation.
Yob, G. (1973). Hunt the Wumpus [BASIC]. People’s Computer Company.