Chapter One

Literature Review

Play and Games

While romantics like Jean-Jacques Rousseau and Friedrich Schiller idealised play as a state of Nature and innocence, Victorian society tended to oppose play to productive industry, presuming a normal and desirable progression from child's play to adult work (Cohen, 1987). Huizinga (1955) and Caillois (1961) established the more modern conception of play as an ongoing, positive force in adult life and, more broadly, human civilisation, but their oft-cited definitions nonetheless emphasise the border between play and other types of activity. For Huizinga (1955), play is:

a free activity standing quite consciously outside ‘ordinary' life as being ‘not serious', but at the same time absorbing the player intensely and utterly. It is an activity connected with no material interest, and no profit can be gained by it. It proceeds within its own proper boundaries of time and space according to fixed rules and in an orderly manner. (p. 13)

Caillois (1961) elaborated upon Huizinga's (1955) arguments to provide a more detailed definition of play as:

• Free: in which playing is not obligatory; if it were, it would at once lose its attractive and joyous quality as diversion;
• Separate: circumscribed within limits of space and time, defined and fixed in advance;
• Uncertain: the course of which cannot be determined, nor the result attained beforehand, and some latitude for innovations being left to the player's initiative;
• Unproductive: creating neither goods, nor wealth, nor new elements of any kind; and, except for exchange of property among the players, ending in a situation identical to that prevailing at the beginning of the game;
• Governed by rules: under conventions that suspend ordinary laws, and for the moment establish new legislation, which along counts;
• Make-believe: accompanied by a special awareness of a second reality or of a free unreality, as against real-life. (Caillois, 1961, p. 9)

Caillois (1961) also divided games into four categories: “agon,” “alea,” “mimicry” and “ilinx.” Agonistic games involve players pushing their skills through competition. Aleatory games are games of chance in which the play tries to forecast or control chance in some degree. Games of mimicry include all forms of impersonation in which one takes on a pretend role. Ilinx games involve an altered state, such as making oneself dizzy, thrilled, or terrified. For Caillois, there was a ‘progression' within each of these types, from the extreme of active exuberance, spontaneity and turbulence, to control, calculation, effort and subordination to rules; the former he termed “paidia,” the latter “ludus.”

While Huizinga (1955) and Callois (1961) saw play as an end in itself, it is now usually seen in terms of its psychological or social functions (Cohen, 1987; Millar, 1968; Yawkey and Pellegrini, 1984). Within the earlier psychoanalytic tradition, play was seen as an expression of repressed desire that (like dreams) dramatised unconscious content in a way that could be useful in a therapeutic context (Klein, 1932; see also Schaefer & Reid, 1986). More recently, in self psychology, play has come to be seen as central to the development of self, notably in terms of early interaction between the mother and child (Schore, 1994; Stern, 1977, 1865), the use of “transitional objects” (Winnicott, 1971), and “selfobjects” (Kohut, 1971, 1977) (also see Meares, 1992; Mitchell & Black, 1995; Stein, 1991).

Within the Constructivist tradition derived from Jean Piaget (1929/1973, 1932/1965, 1950/2001, 1951, 1978), play is often seen as a form of rehearsal for sensorimotor and cognitive skills. Piaget (1951) distinguished between three types of play. “Practice play” is associated with the repetition of actions as a means of mastering sensorimotor skills. “Symbolic play” is associated with imitation, fantasy and socio-dramatic processes. At this stage play is often egocentric, in the sense that it does not involve true co-operation or the consistent application of rules. Piaget's third category is “constructional games” with prescriptive rules, including hide-and seek, marbles and board games. At this stage, rules are seen as fixed at a point in time through mutual consensus and are changeable through a renewed consensus. Piaget's (1951) model therefore describes a developmental progression from rule-less egocentric play to rule-governed, co-operative games, paralleling Caillois' progression from paidia to ludus.

Taxonomies have been produced which identify the different ways that games structure the relationships between players. Avedon (1971), for example, identifies ten structural elements of games: purpose, procedures, rules, number of players, roles of participants, results or pay off, abilities and skills required, interaction patterns, physical setting and environmental requirements, and required equipment. Redl, Gump and Sutton-Smith (1971) go further, providing a taxonomy of thirty dimensions of games, among them bodily contact, bodily activity, skill requirements, the role of chance and competition, the use of space, time and props, role-taking factors, and role complexity, with subcategories. However, within anthropology games are often seen in relation to rituals as microcosms of social structures and as a site where such structures may be reproduced (Geertz, 1973; Mead, 1934/1972). Certainly, researchers have become increasingly concerned with the relationship between video games and learning. It has been argued that video games improve spatial and cognitive skills (Braun & Giroux, 1989; Greenfield, 1981; Greenfield et al., 1994; Kinder, 1991; McClure & Chaille, 1987; Subrahmanyam and Greenfield, 1994), and Seymour Papert (1980) has argued that games may function as “microworlds” that facilitate children's learning. Indeed, the growing awareness of the relationship between cognition and emotion has meant that theories of children's and adult's education now tend emphasise the motivating role of the positive emotional states fostered by play (Kafai, 1995, 1996; Malone, 1981). In this respect, play and games are seen as a socialising force as well as an instrument of pedagogy.

Recently, some researchers of video games have suggested the need for a discipline of “ludology,” or the formal study of games (Eskelin, 2004; Frasca, 1999, 2003). This notion of ludology as a discipline is a timely response to the prevalent narratological and dramatic models of gameplay, and seems logical. After all, play is defined as separate from other activities, and may be experienced as separate from non-play activities. However, one must be wary when play's separateness is over-emphasised. If we do not consider every system of rules a “game,” and almost any activity may be engaged with “playfully,” it can be difficult or misleading to separate play from other processes. As is evident in Bordwell's (1986) analysis of film, and Grodal's (1998, 2003) analysis of film and video games, the cognitive or affective work engaged in an aesthetic experience are not reserved for that experience, they are mediated versions of everyday cognitive and affective activity.

Cognitive Activity in Art and Games

In accounting for the cognitive skills of video game players a useful departure point is Bordwell's (1986) and Branigan's (1992) analyses of film, which focus on inferential perceptual-cognitive activities. For Bordwell (1986):

Perception [is] a process of active hypothesis testing. The organism is tuned to pick up data from the environment. Perception tends to be anticipatory, framing more or less likely expectations about what is out there. . . . The organism interrogates the environment for information which is then checked against the perceptual hypothesis. The hypothesis is thus either confirmed or disconfirmed; in the latter case, a fresh hypothesis tends to appear. (p. 31)

This inferential activity is organised by “schemata,” relatively constant hypothetical structures that classify and organise information, providing a basis for expectation, and Bordwell differentiates between “prototype,” “template,” and “procedural” schemata. Prototype schemata “involve identifying individual members of a class according to some posited norm,” and often involve the identification of “agents, actions, goals, and locales” (p. 34), or “identifiable persons, actions, locales” (p. 49). They are, then, readily identifiable as types, or stereotypes. Template schemas are larger structures that “can add information when it is absent and test for proper classification of data” (p. 34). Bordwell argues that the “canonical” story is a template schemata comprised of: “introduction of setting and characters—explanation of a state of affairs—complicating action—ensuing events—outcome—ending” (p. 35). Template schemata may, then, be seen as analogous to the codes and conventions of particular genres. Procedural schemata, by contrast, are those “operational protocols which dynamically acquire and organise information” (p. 36). Bordwell uses the term to refer to higher-level cognitive processing. For example, if material in the syuzhet does not conform to canonical schemata the viewer may turn to “compositional,” “realistic,” or “transtextual” schemata to interpret them, “search[ing] for appropriate motivations and relations of causality, time, and space” (p. 49) which make the material intelligible.

A film, then, “presents cues, patterns, and gaps that shape the viewer's application of schemata and the testing of hypotheses” (p. 33). The viewer proceeds by using a mixture of “bottom-up” processing, making inferences “on the basis of perceptual input,” and “top-down” processing, organising perceptions on the basis of expectations drawn from known schemata (p. 31). The viewer thereby organises “chunks” of film into “more or less structurally significant episodes” (p. 35). In this respect, the “artwork is necessarily incomplete, needing to be unified and fleshed out by the active participation of the perceiver” (p.32), but the artwork sets limits on the activity of the spectator, encouraging the application of certain schemata. What is significant is that perception when engaging with an artwork differs from other forms of perception:

In ordinary perception . . . perceptual hypotheses tend to be vague and open-ended. In art, however, alternative hypotheses tend to be much more explicitly defined, their set tends to be closed, and they get challenged fairly often . . . Narrative art ruthlessly exploits the tentative, probabilistic nature of mental activity. (p. 39)

Consequently, Bordwell foregrounds the hypothetical activity of the viewers in terms of curiosity and suspense, arguing that: “narratives are composed in order to reward, modify, frustrate, or defeat the perceiver's search for coherence” (p. 38).

Grodal (1997) argues that, compared with cognitive models, structuralist models are too “static,” presuming structures within texts, while poststructuralist accounts tend to separate meaning from its adaptive foundations in the individual (p. 14). However, language provides a sequential way of representing the holistic experience of thought, and remains an ongoing determining factor in planning, evaluating, remembering and reasoning (Vygotsky, 1934/1962). In this respect, the accounts of textuality in Cultural Studies offer a sensitive terminology that can be usefully integrated with cognitive models (see Hernadi, 2002; Richardson & Steen, 2002; Spolsky, 2002; Turner, 2002). Indeed, as Stam (2000) argues, cognitivism can be viewed “as a nostalgic move backward” (p. 240) in the sense that “a focus on cognitive commonalities across all cultures exists below the threshold of cultural and social difference, and therefore discourages analysis of tensions rooted in history and culture” (p. 242). Analysing cognition tends to locate meaning within the viewer, making it easy to forget that schemata, like codes, are acquired in a social context and may define an audience's shared culture.

Bordwell's (1986) account is also limited because the cognitive work of video game players is bound up in sensorimotor activity. Indeed, while Bordwell uses the term “procedural schemata” to refer to abstract cognitive work, the term usually refers to the distinction between, on the one hand, an episodic or semantic understanding of an action (“declarative knowledge”) and, on the other hand, knowledge which is implicit in performance, including the registering of sensory cues and responding to the present environment (“nondeclarative knowledge”) (Squire, 1987). In this respect, procedural schemata are associated less with understanding than behaviour, and are of particular relevance to video gameplay, which has a performative component.

Turkle (1984, 1995) may be seen as accommodating the motor activity of computer users when she differentiates between “abstract” and “concrete” styles of computer use. The abstract style, epitomised by structured program design, may be characterised by a preponderance of top-down processing, in that users have a model of what they want to do at the interface then perform whatever sensorimotor activity is required to carry it out. The concrete style, epitomised by an unstructured, bottom-up approach to program design, may be characterised by a preponderance of bottom-up cognitive processing and ongoing sensorimotor activity, in that the user learns through experimentation. However, while Turkle acknowledges that these two styles may be mixed, she is more concerned with broad cultural styles and aesthetics than with the real-time dynamics of motor and cognitive activity.

Loftus and Loftus (1983) provide a more useful foundation of the dynamic relationships between cognitive and motor activity in video games. They observe that the mind and body are not unitary systems, but depend upon shared components and processes: sensory memory, short- and long-term memory, attention, reaction time, cognition, and the motor system. Players constantly monitor an original state, a goal state, and the means for achieving the latter, but while several strategies are often available to achieve a goal each strategy has its advantages and disadvantages. For example, a player who relies on their long-term memory of a game environment must continually remember where they are in a given route, will be disoriented when they take a wrong turn, and will have “less processing capability for such things as focusing and switching attention” (p. 71). However, players can choose strategies which work for them. A poor memoriser may choose a strategy that requires little memorisation, while someone who takes too long to retrieve long-term memories may choose strategies that minimise the requirements of such retrieval. In this respect, gameplay may be analysed in terms of the cognitive and sensorimotor strategies that players use to achieve their goals in the game.

The Aesthetics of Video Games

Game aesthetics can be usefully distinguished by accounting for the cognitive and/or sensorimotor demands they make of players and the extent to which these make play more engaging. Malone (1984), building upon Papert's (1980) work, identified four categories of elements which he argued were characteristic of “intrinsically motivating” games: first, the existence, clearness, accessibility and meaningfulness of goals; second, the importance of uncertain outcomes through the use of variable difficulty levels, multiple goal levels, randomness, and hidden information; third, an emotionally appealing fantasy that was intrinsically related to skills learned and provided a useful metaphor; and, fourth, curiosity, both sensory (audio and visual effects, as decoration, as enhancing fantasy, as a reward, and as a representation system) and cognitive (surprises and constructive feedback).

Malone's research has been influential, and is useful from both a design point of view and as a checklist for analysis, but his taxonomy is insensitive to players' aesthetic preferences. Myers (1990c), by contrast, developed Malone's work, using questionnaires (Q-methodology) with a sample of 44 gamers, to identify dominant aesthetics in the subjects' favourite video games. As the basis for his questions, Myers uses four criteria: challenge, curiosity, fantasy and interactivity (and their opposites: ease of play; realism; familiarity; and stability/constancy). He sorts players' responses into five principal “factor arrays”: the game as a new challenge (skill building); the game as a pleasant, non-threatening social diversion; the game as meditative withdrawal; the game as an enemy; and the game as a new challenge (skill-proving). He suggests that the three most common aesthetics are: the “game as challenge, subcategorised as (a) game providing a challenging opponent (game as referee) and (b) game being a challenging opponent (game as enemy)”; the “game as meditation”; and the “game as social activity (more common in video than home computer games)” (p. 385).

The dominant aesthetic in the above accounts are based on skill mastery, and certainly:

many investigators have concluded that the appeal [of video games] rests in challenges to improve one's skills, to attain a higher level of game complexity, to persist in the face of obstacles and setbacks, to exceed one's own or some other target of excellence. (Durkin, 1995, pp. 16-7)

A sense of development, accumulation and mastery is evident in many elements of game structure: the use of high scores, levels, quests and bonuses; and step-by-step increases in pace, the difficulty of opponents and levels, and the power of characters. These intrinsically motivating aspects of gameplay are amenable to Wallace's (1999) use of “operant conditioning,” in that the placement, timing and scale of problems and rewards determine the continued interest of the player.

More generally, Turkle (1984) argues that young children quickly develop concerns with “domination, ranking, testing, proving oneself” (p. 58) whose roots are “aggressive, passionate and eroticized” (p. 59). In this respect, the tendency towards skill development and mastery may be related to the predominantly male demographic of video games (Anderson & Ford, 1987; Cacha, 1983; Chambers & Ascione, 1987; Cooper & Mackie, 1986; Dominick, 1984; Eysenck & Nias, 1978; Graybill, Kirsch & Esselman, 1985; Kaplan, 1983; Kaplan & Kaplan, 1981; McClure & Mears, 1984). Indeed, Skirrow (1990) uses Melanie Klein's (1932) psychoanalytic theories about the relationship between infants and their mother's bodies to argue that gameplay is a masculine discourse that includes:

a reactivation of infantile feelings triggered by the games, and, partly, perhaps, the attractions of masochism. . . . Continuing to perform in the game . . . is the same as continuing to live. When you stop you die” (p. 331).

For Skirrow, an internalised anxiety about an unreal danger, understood in terms of castration anxiety, is externalised with the embodiment of the “penis-as-magic-wand” (p. 331). Kinder (1991) similarly argues that “the repetitive, segmented, serial nature” (p. 110) of game narratives, during which players are “constantly threatened by short circuiting and premature deaths” (p. 111), “leads to a disavowal of obsolescence, castration and death” (p. 110). For both Skirrow and Kinder, then, the increasing mastery of the typically male protagonist is a pre-emptive attempt to protect the male ego, or phallus, from annihilation or diminution.

Turkle (1984) argues that young players appreciate the consistency of the strictly “rule-governed worlds” (p. 77) of games, and in many respects a game functions as an opponent whose strategies are gradually learnt and mastered. Myers (1990a) and Friedman (1995), following game designer Chris Crawford (1984), argue that video games reveal their constructedness more than other media. Friedman (1995) observes that:

Learning and winning . . . a computer game is a process of demystification: One succeeds by discovering how the software is put together. The player molds his or her strategy through trial-and-error experimentation to see “what works”-which actions are rewarded and punished. (p. 82)

In short, video games, being “simulated, rule-driven worlds” (Turkle, 1984, p. 74), allow players to discover, or demystify, their inner workings, rules, “secrets,” and “logic structures” and, ultimately, how they are “put together” (Kirksaether, 1998, p. 82). Indeed, Wertheim (1999) argues that:

[the adventure game] Adventure may be seen as a metaphor for computing itself. During the game, players cracked the code of this virtual world in much the same way that a hacker would crack the code of a computer operating system. (p. 246)

Herz (1997) similarly observes that: “charting out subterranean passages and dead ends is pretty much analogous to mapping out a circuit or debugging a piece of code (p. 11). Unlike print or film texts, then, a game may be replayed until the structure of the software has been apprehended, and manuals, magazines, and other game accessories help players expose this structure.

Nonetheless, Friedman (1995) overstates the importance of “demystification” when he argues: “the moment [a game] is no longer interesting is the moment when all its secrets have been discovered” (p. 82). He implies one over-riding gaming genre, style or aesthetic of (explorative) mastery, when games are inherently repetitive and allow for replay. Object-event sequences, including enemy activities, are often randomised, and in multi-player games the demystification necessarily extends to the strategies of another player. Unless one is capable of plotting in real-time all factors that determine randomisation in a video game, or another player's strategy is completely understood and predictable, total mastery is impossible. Even Malone (also see Provenzo, 1991 and Wallace, 1999) observes that game appeal is influenced by the degree of randomness, and Chandler (1994) writes:

most good games are not solely a test of skill, but offer an element of chance which can encourage those who less confident of their current level of playing skill. Children, I would suggest, are well aware of this feature, and many prefer games in which they can blame luck for their failures. (¶ 18)

So while some players like “testing their worth” against an unforgiving machine, others dislike the rigidity of strictly rule-governed domains, finding it an “intolerable. . . pressure, . . . a taunt, a put-down” (Turkle, 1984, p. 86).

Competition between players may provide a context for challenge and mastery, but the social dimensions of video games may constitute a distinctive aesthetic. Provenzo (1991) argues that arcade gaming constituted a sub-culture that was one of the “great equalizers of youth culture because [they] allow[ed] an eight year-old to approach a fifteen-year-old and discuss something as peers” (p. x). Other research (Lepper, 1985; Mitchell, 1985) has indicated that video games may promote filial interaction when youth play within the family environment and talk about games with their parents. Unfortunately, early arcade culture was lost with the transformation of arcades into sanitised places where parents were encouraged to leave their children in shopping malls (Herz, 1997). However, games still may be played with friends or family; shared knowledge of video games still provides a basis for social interaction; and Local Area Network (LAN) parties and game conventions are contemporary spaces for collective gameplay. The problem is that much of the recent research on computer-mediated communities has tended to focus on specialised PC-users, notably those who use Multi-User Dungeons (or MUDs) (see Baym, 1995; Beaubien, 1996; Chesebro & Bonsall, 1989; Curtis, 1996; Dery, 1994; Foster, 1997; Jones, 1995, 1998a, 1988b; Kim, 2000; Kolko & Reid, 1998; Kollock & Smith, 1999; Porter, 1997; Stone, 1995; Rheingold, 1991; Strate, Jacobson & Gibson, 1996; Wilbur, 1997).

Video games may also serve a social function without other players. Selnow (1984) has argued that children may regard the game itself as “a kind of surrogate companion and their interactions with the games as social interactions” to the extent that the game becomes an “electronic friend” (p. 58). Turkle (1984) argues that video games offer a form of useful escape for disabled children, and online interaction especially allows disabled children social interaction otherwise denied them. Wallace (1999) makes a more general argument that it “is hard for any humdrum reality to compete with [the worlds of video games], especially for people whose lives are troubled by low self-esteem, boredom, lack of social support, or unsatisfactory personal relationships” (p. 172). Video games, then, offer a compensatory sense of achievement, autonomy and self-esteem (Greenfield, 1984; Nelson & Carlson, 1986; Turkle, 1984). However, some children may feel “cut off” when they finish a game (Turkle, 1984, p. 66), and it is possible that too much control over a game world may lead children to be impatient with a less obeisant real world (Sheingold, cited in Greenfield, 1984, p. 114).

While mastery and social interaction may play important roles in the aesthetics of gameplay, the experience of identification, immersion and agency is of equal importance. The departure point for early theorists concerned with these experiences was television (Durkin, 1995, 1999; Greenfield, 1984; Turkle, 1984). Greenfield (1984) argued that both TV and video games have a visual appeal, and that their moving images especially appeal to young viewers (p. 88-90). Indeed, for her, familiarity with TV is one reason why children respond so readily to video games. For Stallabras (1996), by contrast, video games are distinguished from other media by their use of interaction: “the passivity of cinema and television is displaced by an environment in which the player's actions have a direct, immediate consequence on the virtual world” (p. 85). However, Greenfield (1984), Turkle (1984), Skirrow (1990), and Kinder (1991) argue that what is truly distinctive about video games is that they combine a passive mode of spectatorship associated with cinema and television with more active modes of participation and interaction. For Skirrow (1990), the “‘audience' disappears as a distinction between ‘doer' and ‘viewer'” (p. 330), and the enigmas for the performer are of the order of “Where am I?” rather than “Who am I?”

Durkin argues that, compared with TV and film, video game players “do not feel a strong sense of identification with the characters” (1999, p. 129), and he leaves open the question as to whether there are less behavioural effects in games. Fuller and Jenkins (1995) similarly note that, compared with print and film texts, identification with characters in video games is limited because characters tend not to have psychological depth:

Characters play a minimal role, displaying traits that are largely capacities for action: fighting skills, modes of transportation, preestablished goals. . . . Activity drains away the character's strength, as measured by an ever shifting graph at the top of the screen, but it cannot build character, since these figures lack even the most minimal interiority. (p. 61)

Skirrow (1990) is even more critical, stating that: “for the performer of a game the first and third person are almost totally identified, so there can be no suspense based on knowing more (having seen more) than the protagonist who represents you” (p. 330). The major objection to such accounts is that the characters in many print and film texts do not necessarily offer more psychological depth than the average video game character. We cannot judge video games solely on the basis of their status as a medium, since a video game with a complex narrative and detailed characterisation may offer more depth than a poorly written novel. In this respect, the above accounts ignore the distinctiveness of video games by evaluating them in terms of other media.

A s Turkle notes, in a video game: “you have to do more than identify with a character on the screen. You must act for it. Identification through action has a special kind of hold” (1984, p. 79). The issue, then, is how “identification through action” creates a distinctive hold. Friedman (1995) observes that in earlier simulations, like Civilization (1991), one identified principally with a leader from a particular culture or historical period, such as Genghis Khan, indicating a straightforward identification with the role of a leader. However, in more recent simulations like SimCity (1989) there is a “constant shifting of identificatory positions depending on whether one is buying land, organizing the police force, paving the roads, or whatever” (p. 85). Yet simply “attempting to map ‘roles' onto the player's on-screen identification misses the point” (p. 85) since:

overarching these functional shifts . . . is a more general state of identification: with the city as a whole, as a single organism. . . .[or] a process. . . . ‘Losing oneself' in a computer game means, in a sense, identifying with the simulation itself. (pp. 84-85)

Friedman concludes that the player does not simply identify with a role:

the player forms a symbiotic circuit with the computer, a version of the cyborgian consciousness described by Donna Haraway (1985) in her influential “Manifesto for Cyborgs.” The computer comes to feel like an organic extension of one's consciousness, and the player may feel like an extension of the computer itself. (1995, p. 83)

The term “cyborg” is, of course, a contraction of “cybernetic organism”: a synthesis of the technological and biological. This is usually understood in terms of the physical augmentation of humans with technology, but in practice a cyborg is any being that forms a cybernetic relationship—a dynamic, self-regulating, and homeostatic (information) feedback system—that incorporates both organic and non-organic material (Gray, 1995). This means that a person need not be physically fused with a computer to be a cyborg—any causal relationship with technology, such as the feedback loops that occur during gameplay, suffices for the term to apply. Therefore Friedman (1995) could be read as simply referring to the psychological experience of this relationship.

The problem is that Haraway (1985) sees cyborgs as subverting traditional boundaries between physical/non-physical, self/body, human/machine, and thereby creating post-, proto-, and/or non-human beings. On this basis she sees cyborgs as a polemical metaphor for the subversion of all binary systems, including male/female, which presume a primary term. Theorists of computer-mediated communities, including virtual reality, sometimes seem to conflate the literal and metaphorical meanings of the term, implying that cyborgs are inherently subversive or revolutionary (see Featherstone & Burrows, 1995; Gray, 1995; Haraway, 1985; Helsel & Roth, 1991; Reid, 1994; Stone, 1995; Strate, Jacobson & Gibson, 1996; Wolmark, 1999). However, to claim that the cybernetic relationship is inherently subversive is to fall back on essentialism, when it should be obvious that a cybernetic relationship may be read in different ways. Players may simply see the game as a tool and experience immersion without any sense that their humanity, or any other aspect of their identity, has been altered. It suffices to suggest that, in either its literal or metaphorical form, the term “cyborg” only offers a limited account of the relationships between players and their games.

Turkle (1986) provides a more useful departure point when she argues that what players pursue in their interactions with video games is an “altered state,” described less in terms of books and TV than by metaphors taken from “sports, sex or meditation” (p. 60):

Call it ‘muscle memory,' call it ‘flow,' call it ‘trusting your instincts' – the experience of feeling a continuity between mind and body is part of the inner game of any well-played sport. Skilled video game players experience this immediacy of knowing their game with more than their head, and the experience is exhilarating. (p. 81)

An account of such altered states can be found in Csikszentmihalyi's (1975, 1990, 1993) description of “flow.” For Csikszentmihalyi (1975, 1990, 1993), “flow” is “the state in which people are so involved in an activity that nothing else seems to matter; the experience is so enjoyable that people will do it even at great cost, for the sheer sake of doing it” (p. 4). Csikszentmihalyi (1993) identifies eight dimensions of flow: clear goals and immediate feedback; a balance of challenge and skill; the merging between consciousness and one's activity; a high degree of focus; the loss of self-consciousness; an altered sense of time; and an autotelic quality. The flow state, then, is facilitated when there is sufficient challenge to stimulate the individual beyond their normal level of effort: too much challenge creates anxiety, too little creates boredom, and in this respect flow may be seen in terms of an appropriate schedule of reinforcement (Loftus and Loftus, 1983; Wallace, 1999). It can be argued that play has been increasingly identified with—or is seen at its most optimal when players enter—a state of “flow.” In this context, the problem with the term “play” is that it often refers to an activity which may be characterised as an ongoing attempt to (re-)enter the ideal of play: a state of flow that is totally absorbing and gratifying to the ego. Furthermore, while it is useful to analyse how games produce flow experiences, the fact that there are different metaphors for “altered” or “flow” states suggests that it may be better to turn away from generalisations about some universal, ineffable state and attempt to distinguish different types, or qualities, of immersion.

As Murray (1997) argues, “immersion” is “a metaphorical term derived from the physical experience of being submerged in water” and what people seek from “a psychologically immersive experience” is:

the sensation of being surrounded by a completely other reality, as different as water is from air, that takes all of our attention, our whole perceptual apparatus. We enjoy the movement out of our familiar world, the feeling of alertness that comes from being in this new place, and the delight that comes from learning to move through it. (pp. 98-99)

Murray traces the distinctive qualities of immersion to the “procedural,” “participatory,” “spatial” and “encyclopaedic” characteristics of the computer as a medium. She argues that the computer is procedural in the sense that: “it was designed not to carry static information but to embody complex, contingent behaviors” (p. 72); in short, it is an engine of computation that operates according to programmable rules. The computer is participatory in the sense that it is “responsive to our input” (p. 74), and, for Murray , the notion of “interaction,” applied to computers, refers to an environment “that is both procedural and participatory” (p. 74).

The spatial qualities of the computer are also distinctive because, unlike “linear media such as books,” the computer does not simply represent space in verbal or pictorial form; it represents “navigable space” that users can “move through” (Murray, 1997, p. 79). Indeed, “the computer's spatial quality is created by the interactive process of navigation” (p. 80), in that computers not only calculate spatial representations from a player's perspective, they do so dynamically, in real-time. Lastly, Murray argues that computers are encyclopaedic, in the sense that they can store so much information, making them more capable of representing a procedural, participatory, and spatial world , not just in terms of visual detail, but also scope and dynamics. This encyclopaedic quality provides a basis for “epic-scale narrative” (p. 84), evident in online role-playing games (p. 86) that allow multiple players to create and share occupancy in a developing game world. For Murray , the combination of the spatial and encyclopaedic elements with the procedural and participatory elements is what provides the aesthetic experiences of immersion and the related experience of “agency,” that is, the sense of entering a navigable virtual space and seeing the consequences of one's actions in the virtual world.

It must be emphasised that while full sensory immersion is the ideal of Virtual Reality (see Heim, 1993), video games only offer limited physical immersion, ranging from the wrestling, slapping and kicking of the joystick or cabinet with arcade games to the less physically taxing interaction with a PC game's keyboard and mouse controls. Darley (2000) argues that most video games offer a heightened form of “kinaesthesia”: “the illusion of control and agency in real time and in a realistic looking (and behaving) environment [italics added]” (p. 154). For him: “deliberation is minimal – once the game is under way the player is compelled continually and immediately to respond” (p. 156). Skirrow (1986) has similarly observed that video games often produce a sense of immediacy and danger that leads to “crisis management” (p. 331), and Turkle (1984) has emphasised the importance of survival, in that, in many games, “one false move [and] you're dead” (p. 79). The low tolerance for error in a context of high stakes fosters an increasing level of attention and a more intense or immersive altered state. In this respect, the reflexive and frenzied character of interaction in many video games may compensate for absent sensory channels. However, video games do not necessarily emulate other media and so cannot be seen as merely compensatory. As Murray (1997) argues, they generate their own aesthetics through creative experimentation with their constraints.

Video Game Content

It is perhaps understandable that early researchers were concerned about the values, or content, of the video games in which players were immersed, and acted. Provenzo (1991), who reflects these concerns, argues that video games are generally sexist and violent. He argues that females, when they are represented at all, are usually ridiculous stereotypes of femininity and serve as objects to be looked at or rescued (see also Cassel & Jenkins, 1998; Consalvo, 2003; Herz, 1997; Poole , 2000; Turkle 1984). Not only are males the dominant demographic, video games appeal to their aggressive instincts. For Provenzo, the “message communicated by the rules of [video games] is that violence is not only acceptable, it is necessary to win” (p. 124). Stallabras (1996) makes a similar claim, observing that video games do not show the consequences of violence: bodies disappear after a period of time; no matter how they die, the corpses of each type of monster look the same; aggression and murder are seen as merely an expenditure of energy that lowers one's health bar; and injuries are regenerated by eating food or camping for a night's sleep.

What Provenzo (1991) finds especially alarming is that many games draw their scenarios from historical events, such as World War I and II, Vietnam , the Cold War, and the Gulf War. That is, when video games do historicize violent conflict, it is usually to harness or reinforce racial or national stereotypes, such as the notion that Chinese or Japanese are different, might want to fight us, and therefore must be regarded as the enemy. There is, certainly, a strong relationship between video games and the military-industrial complex (Herz, 1997; Stallabras, 1996; Toles, 1985), and many games draw their iconography from warfare: arsenal, field armour, ammunition, health kits, and military-style missions, maps and strategy.

More recently, Fuller and Jenkins (1995) have argued that mastery of the game world gains symbolic value from the New World narratives of Renaissance explorers (1995, p. 58). Players expand their wealth and power, and bring space “under symbolic control” (p. 69) by continually pushing back or establishing such frontiers as the edge of the screen, save points and levels (p. 67). Using Michel de Certeau's (1984) argument that narrative involves the transformation of space into place, Fuller and Jenkins (1995) suggest that the rhetoric of games as new frontiers or New Worlds is driven by:

the desire to recreate the Renaissance encounter with America without guilt: This time, if there are others present, they really won't be human . . . or, if they are, they will be other players like ourselves, whose bodies are not jeopardized by the virtual weapons we wield (p. 59)

Kinder (1991) and Stallabras (1996) link gameplay to the conservative values of capitalist exchange. Kinder (1991) draws from Piaget (1978) and Lacan (1977/2001) to argue that video games prepare children for consumer culture by fostering interactivity in the context of multi-media merchandising networks. Stallabras (1996), who is more polemical, argues that video games:

obsequiously reflect the operation of consumer capital for they are based on exchange, an incessant trading of money, munitions or energy, a shuttling back and forth of goods and blows. (p. 90)

For Stallabras, video games are a debased “culture industry” that distract the “masses” from larger social issues such as global inequality.

However, there has been a trend in the industry to design games for girls, reflected in Laurel 's ill-fated Purple Moon company and the Grrlz Games Movement (see Cassel & Jenkins, 1998). There have also always been a few games, including Final Fantasy , that have dealt with social issues such as feminism, economics, ecology, science and religion. More to the point, setting aside critics' evaluation of the moral or political values of games, players' perception and acceptance of a game's values is not a simple issue.

Subjectification and the Play of Meaning

For both conservatives like Provenzo (1991) and polemicists like Stallabras (1996), what distinguishes video games is not merely their undesirable values, but the extent to which players participate in them. Their arguments therefore may be said to rest on the assumption that interaction increases not only identification and immersion, but also subjectification. There is, it must be acknowledged, a rudimentary logic to this. As Andrew Darley (2000) observes, video games are increasingly characterised by “the impression of being enabled to act within and upon the world one gazes upon [italics added]” (p. 161). That is, while video games are defined by their use of interactivity, they offer “a kind of relative or regulated agency: the constraints of the game allow the player to choose between a limited number of options [italics added]” (p. 164). This becomes significant when the choices players are forced to make require them to act out certain types of values. As Herz (1997) observes, Will Wright built certain values into SimCity (1989) in the sense that the rules of the game promote certain types of actions, such as the use of public transport. Winning the game is easier when one conforms to those rules, and such conformity could be interpreted as acceptance.

However, despite its seeming common-sense, the assumption that interaction facilitates identification recapitulates an outmoded notion of identification. Within early film theory, which drew from Althussurian Marxism and Lacanian psychoanalysis, the more one identified with a film or a character the more vulnerable one was to the film's ideological position (see Lewis, 1984; McGree, 1997; Metz , 1975, 1982; Nichols, 1981; Palmer, 1989; Penley, 1989; Staiger, 1992). Cultural Studies no longer accepts this position, but it nonetheless persists. For example, Stallabras (1996) has claimed that interaction in video games is so routine and mechanical that players subjectify themselves as mindless object-machines beneath the malevolent forces of late capitalism. In a less polemical account, Taylor (2003) draws on psychoanalytic film theory to argue that breaks in point of view fracture identification, thereby recapitulating the notion of the cinematic apparatus as the basis for identification at the very moment she sees the video game as offering a different kind of apparatus.

While the relationship between “interactivity” and “subjectification” is still open to debate, recent research has begun to address the issue in more subtle ways. Frasca (2003) identifies four levels of ideological operation in “simulations,” including computer games. The first level, shared with film and narrative in general, occurs at the level of representation, and involves the signification of the “objects and characters, backgrounds, settings, and cut scenes” (p. 232). The second level relates to “manipulation rules: what the player is able to do within the model” (p. 232). As Consalvo (2003) has observed, the provision for certain types of actions may naturalise or promote an ideology of sexism in FFVII : skills such as “Protect Girl” reinstate traditional gender relations as part of the text. Third, there are “goal rules,” which define actions that are not merely possible , but which the player must perform if s/he is to win. The fourth level deals with “meta-rules,” when the player is able to change the rules. We might include the use of hints, spoilers, and cheats in this category.

Theorists could distinguish more complex causal conditions across the spectrum of possible and necessary actions, for example in the manner suggested by Eselinen and Tronstad (2003). However, this thesis takes the position that performance of an action in a video game does not necessarily entail endorsement of that action, and that players have the capacity to play not only in the sense of making physical choices, but also in the sense of playing with the meanings of those choices. This position reflects the changing ways that Cultural Studies has appropriated the terms “play” and “game” to describe the act of reading. Within everyday discourse, as in the phrase “to play a game,” “play” is a verb that refers to an activity, or a way of doing something, while “game” is a noun that refers to a system of rules. This usage parallels the distinction between parole and langue , as is evident in Lapsley and Westlake 's (1988) summary of structuralist theory:

The system, known as langue , and actual or potential utterances, parole , may be compared to the rule system of chess and to the set of moves that may be actually or potentially played. Langue defines both what are permissible or impermissible utterances (as do the rules of chess in relation to moves) and what their significance is (again, as in chess). . . . If the meaning of a signifier is analogous to the value of a piece on a chessboard, then it becomes evident that meaning will change according to the context in the same way that the value of, say, a pawn will depend upon what stage of the game has been reached, where it is in relation to other pawns, how many pieces are left on the board, and so on. (p. 33-4)

In a structuralist approach, of course, a text's structure is analysed to identify the possible permutations of meaning, such that the “play” of meaning is seen as an “effect” of the text's structure. This constitutes poetics in the traditional sense in that it emphasises the ways in which the reader should read the text (or play the game) and/or the effects that should be produced (such as surprise). Hutchinson (1983) provides a valuable example of this tradition, identifying how authors “signal” a game, arouse readers' expectations, and challenge their deductive and reflexive powers by exploiting a game-text as an “enigma,” a “parallel,” or by utilising such narrative techniques as allegory, allusion and symbolism.

However, in more recent poststructuralist accounts the reader's parole (read: play) potentially distorts and/or transforms the langue (read: game) in and by which it is expressed (and constrained). While this kind of usage is evident in some of Barthes' work (1975a, 1975b), it is consolidated in reader response theory (Holland, 1975; Iser, 1974, 1978), psychoanalytic accounts of the speaking subject (Kristeva, 1980), and deconstructionism (Derrida, 1970), all of which emphasise the open-endedness of signification. Most recently, the subversive quality of “play” characterises much writing on metafiction, the fantastic, and postmodernism (Hutcheon, 1988; Jackson, 1988; McHale, 1987), with a sense that media-savvy producers and audiences exploit semiosis through the manipulation of conventions (see also Bolter, 1991; Landow, 1992). Within Cultural Studies' psychoanalytic framework such playfulness is readily seen as indexical of the subject's desire, which is privileged as a liberating force of resistance against imposed meanings and subject positions. That is, since subjectivity is provisional upon the “rules” of language, our “real” identities (as subjects) may be seen as a form of “role-playing,” as merely a “game.” The term “play” therefore connotes not only the freedom to interpret texts in countless ways, but a changeable state of affairs in which one can “play” a different role in the world beyond the text, and, by breaking the rules, come up with new roles.

Unfortunately, such metaphorical, and often rhetorical, usage is problematic. Even if play is read as indexical of desire, desire is not inherently resistant. As Foucault (1979) argues, desire is never simply repressed by social forces, but is produced within discourses. An analogous conclusion has been reached in the psychoanalytic revision of fantasy as the “mise en scene of desire” (Laplanche & Pontalis, 1988, p. 318). Donald (1989) summarises:

What is evident in fantasies is not just the wish, but the setting out of the wish in a way that always also incorporates the prohibitions, censoring, and defences that surrounded the wish. (p. 140)

That is, desire always operates within a social context, and its expression may reinscribe the prohibitions which gave rise to it. By analogy to play and games, sometimes the point of a game is not to win, but simply to prolong the experience of play. Even when one desires to win, and cheats, this may legitimate winning as worthwhile, since cheating is usually an expression of a desire for the ideal subject position offered by a game. In short, the desire expressed when cheating may sometimes be less a violation of the game itself than a violation of the rules which govern one's position in the game.

Nonetheless, the notion of a playful reading usefully foregrounds the freedom players' have in producing negotiated and oppositional readings of whatever actions they choose or are forced to perform in a game. Indeed, Jenkins (1992) has argued that individuals can recuperate their existing values by poaching, and reworking, the resources of cultural artefacts. Ultimately there is no reason why players cannot approach the meaning or experience of gameplay in the same ironic or kitsch way that viewers watch B-grade horror films, in which case submission to the limited rules of a game is little different from submitting to a clichéd plot. The problem is that to understand the “interactive” qualities that distinguish video games from other media, and, by implication, to appreciate their distinctive aesthetics, it is necessary to interrogate the border between narratives and games in a literal sense.

From “Interactive Narratives” to “Cybertexts”

Herz (1997) observes that in early video games narrative content functioned as a backdrop for the action or was limited to introductory text and “canned” (pre-filmed) sequences, usually at the start and end of each level (also Juul, 1999; Poole, 2001). These sequences could be seen as “cute or funny but [were] basically irrelevant to their play” (Turkle, 1984, p. 66). However, more recent researchers have begun to identify the tensions, or relationships, between narratives and games in more detail (Buckles, 1985; Juul, 1999; Kirksaether, 1998; Kucklick, 2003; Poole, 2000; Ryan, 2001; Wolf, 2001). Juul (1999), notably, has argued that games and narratives are “two separate phenomena that in many situations are mutually exclusive” (p. 1), concluding that the temporal emphasis on the past conflicts with the emphasis on the present in video games, that the linearity of narrative conflicts with the non-linearity of video games, and that the position of the narrator conflicts with the position of the player.

Nonetheless, Juul (1999) expects that future games will better integrate narratives and games, and Aarseth (1997) has already argued that while there may be differences between the categories of narrative and gaming: “the difference is not clear-cut, and there is significant overlap between the two” (p. 5). This is evident in Darley's (2000) argument that the events of gameplay—opening doors, crossing the screen, picking up objects and so on—can be recounted to construct narratives of “near-epic proportions” (p. 152); or, as Herz contends: “the choreography of an arcade game is its plot” (1997, p. 140). Of course, as Darley (2001) observes, the events which constitute narratives of this sort are impoverished and fragmented when compared with other narrative forms in that they are repetitive, lack closure, have little psychological or narrative depth, and are interrupted by frequent saving, death, and replay (p. 152). In respect to the events of gameplay: “codes that are central to narrative – codes that produce a certain plenitude, a depth of richness and meaning – are largely absent” (p. 153). Darley concludes that “what counts far more [than narrative] is the actual playing , and this involves a certain kind of kinaesthetic performance that becomes almost an end in itself” (p. 151). For him, “narrative in video games is “ de-centred . . . in a subordinate position within the overall formal hierarchy that constitutes the game aesthetic” (2000, p.151).

Yet Darley's (2000) account is true of some games more than others. Some researchers have redeemed adventure and role-playing games in literary or artistic terms, arguing that video games may have the structural or narrative complexity of print and film texts (Buckles, 1985; Kelley, 1993; Randall, 1988; Ziegfield, 1989). The problem is that, as Aarseth (1997) contends, whether or not adventure games make good novels or not is irrelevant, since adventure games are not novels, they are aesthetically distinctive and must be played, and studied, on their own terms (p. 107). Darley (2000) himself acknowledges that “the particular character of interactivity” means that researchers still need to address “what is happening to familiar forms of narrative and [if] other modes and genres [are] involved” (p. 195). In many games, notably interactive narratives, adventure games, and role-playing games, narrative is not necessarily de-centred but central to the aesthetic experience (at least during certain sequences), having been incorporated or reworked in varied, complex ways.

In accounting for the complexity of “interactive narratives” some researchers have used hypertext as a model (see Barret, 1989; Delany & Landow, 1991; Kirksaether, 1998; Landow, 1992; Snyder, 1996). As Ted Friedman (1995) summarises, hypertext:

connect[s] the oppositions of “reader” and “text,” of “reading” and “writing,” together in feedback loops that make it impossible to distinguish precisely where one begins and the other ends . . . calling into question the very categories of author, reader, and text. (p. 73)

That is, whereas a traditional narrative is characterised by a beginning, middle and end, and a predetermined relationship between a writer and reader, those who navigate hypertext and other “interactive” narratives construct their own texts in a non-linear fashion. This has been seen in positive terms as providing a useful metaphor for poststructuralist notions of intertextuality, in that, instead of offering one dominant narrative, voice, or truth, hypertext allows for multiple narratives, voices and truths (Barret, 1989; Delany & Landow, 1991; Jones & Spiro, 1995; Landow, 1992). Indeed, Laurel (1991) has argued that the problem with “interactive narratives” is precisely that players may choose from multiple paths, but that some of these paths may not lead to a dramatically satisfying conclusion. For her, then, the issue is how to program a story that is dramatic in Aristotle's (trans. 1965) sense. However, Aarseth (1997) argues that placing the programmer/computer as a “playwright” simply constructs the user as “both as an agent without a will and as a watcher without a say” (p. 140); that is, Laurel (1991) reduces interactivity to a player's submission to the dramatic decisions of the “playwright.” For Aarseth (1997), the user is never simply an author, and, as Friedman (1995) argues: “the constant feedback between player and computer in computer games is far more complex than [a] simple networking model” (p. 74).

Consequently, after reviewing existing theories of “interactive” media (Anderson, 1990; Anderson, Holmqvist & Jensen, 1993; Eco, 1981; Ziegfeld, 1989), Aarseth (1997) offers his own model. To begin with, he distinguishes between potential sign sequences in a medium, which he calls “textons,” and the signs that actually appear at the surface of a medium, which he calls “scriptons.” He then argues that media can be distinguished in terms of how scriptons determine the textons. In a book the textons and scriptons are identical (the words on the page), whereas with film the textons (the images on celluloid) highly determine the scriptons (the images projected on a film screen). With computers, the textons (registers that store data) are mediated by the layers of hardware and software as well as by the choices of the user before being represented as scriptons (words and images at the interface).

Aarseth (1997) uses the term “ergodic,” derived from the Greek words ergon (work) and hodos (path), to refer to any text in which the activity of a “reader” (partly) determines which signs appear on the surface of the medium. That is, an ergodic text allows the reader to physically select from, alter, or input text, producing a different text that is read from . Aarseth (1997) offers seven variables as a means of classifying the mode of traversal, or ergodic distinctiveness, of texts: dynamics; determinability; transience; perspective; access; linking; and user-functions. Upon this basis, Aarseth differentiates hyperfiction, adventure games, cyborg-authored texts, and MUDs as examples of ergodic literature. However, Aarseth does not see new or digital media as inherently more ergodic than other media, since so-called “linear” media may allow for as much random access as so-called “non-linear” media. The reader of a book may skip sections, read the end first, or consult the index before referring to an earlier section. Certain print texts, such as Choose Your Own Adventure , take advantage of this by making readers choose from multiple narrative sequences, moving back and forth through the physical text. A work of hyperfiction that only allows the user to click forward through a fixed sequence of linked pages is, by contrast, more “linear.” Indeed, to speak of a “text” (reading) as a linear sequence of signs forgets that all nonlinearities are subsumed to the supposed “linearity” of the act of reading (or may be mentally reordered after the “text” is read).

Aarseth's (1997) notion of “text” “is closer to the philological (or observable) work than to the poststructural (or metaphysical) galaxy of signifiers” (p. 20). While readers of such non-ergodic texts as traditional film and print fiction perform the work of reading “in their heads,” the user of ergodic texts:

performs in an extranoematic sense. During the cybertextual process, the user will have effectuated a semiotic sequence, and this selective movement is a work of physical construction that the various concepts of “reading” do not account for. (p. 21)

For Aarseth, early writers on hypertexts confused the playfulness of reading the “text” with the physical playfulness of physically altering the “work,” but in correcting this emphasis he necessarily over-emphasises the “work.” Consequently, while Aarseth's model of the textual-machine specifies a human operator, it is useful to distinguish this from textual-machines that operate themselves and textual-machines with a human reader (see Figure 1.1).

With textual-machines that operate themselves the “work” stands by itself without any accompanying “text.” In the case of textual-machines with a human reader any attempt to understand the internal working of the computer requires that the computational processes (“works”) be accompanied by a “text” dynamically (re-) constructed by the operator. This may include not only a textual model of the computational processes, but a text in the more conventional sense of an interpretation of the “meaning” of the text that appears at the interface. In the case of textual-machines with a human user the “work” is accompanied by a “text” of the “work” as well as a “text” of the cybertext's “meaning,” and both determine the choices the player makes, and how, as a result, s/he alters the “work.”

The relationships between the “work” and the player's model or “text” of this work is implicit in Aarseth's (1997) distinction between “aporia” and “epiphany” in hypertext fiction:

In contrast to the aporias experienced in codex literature, where we are not able to make sense of a particular part even though we have access to the whole text, the hypertext aporia prevents us from making sense of the whole because we may not have access to a particular part. (p. 91)

That is, hypertext readers may have a sense that they cannot make sense of the whole “text” because of absent parts of the “work.” With epiphany, by contrast, “sudden revelation . . . replaces the aporia, a seeming detail with an unexpected, salvaging effect: the link out” (p. 91). For Aarseth:

this pair of master tropes constitutes the dynamic of hypertext discourse: the dialectic between searching and finding typical of games in general. The aporia-epiphany pair is thus not a narrative structure but constitutes a fundamental layer of human experience, from which narratives are spun. (pp. 91-2)

So while the “text” constructed from the “work” may be experienced as incomplete because of absent aspects of the “work,” closure may be experienced through an “epiphany” which seems to unite the “text.”

In fact, relationships between guided speculations (the multiple “texts” cued by the “work”) and the actual outcome (the subsequent “text”) are central to the dramatic experience of narratives and games in general. The pleasure of reading a detective novel, for example, is dependant upon a disparity between the reader's theory (“text”) about what happened and the actual explanation that is finally offered (the completed “work”). This disparity is continually exploited and redressed through the dissemination of clues, red herrings, dead ends, revelations, new events, and so on, which deliberately mislead the reader about “what happened” until the entirety of the “work” is complete. In gameplay, of course, players are not usually concerned with the entire “computational work,” merely those elements that are functionally relevant to play, and different players utilise different aesthetics to determine what is functionally relevant. The frustration consequent upon disparities between “computational work” and “computational text” is only a problem when it exceeds the frustration deemed acceptable according to the aesthetic by which gameplay is evaluated. That is, players do not demystify the computational or algorithmic entirety of the code, data, or medium, they only produce a “computational text” that is adequate for the task of playing the game according to their chosen aesthetic.

Aarseth's (1997) terms allow for a more subtle and sensitive appreciation of video games, but Aarseth is less concerned with elaborating his taxonomy than he is with discussing existing genres. This provides for some useful insights into the structural differences between commonly accepted genres, but a detailed map of all the ergodic features of a genre or text is less important than the player's model of its sign system, or, rather, the tensions between a player's expectations about feedback and the feedback they actual receive. Given that this study is largely concerned with players' experiences, it avoids referring to ergodics in Aarseth's sense of the game “work,” instead referring to “ergodic codes” that define admissible choices for interaction at the interface. This allows us to separate out conventions of an interface that may characterise a genre, but which are portable between genres. Like all codes, ergodic codes are usually invisible when the user is familiar with the interface, but are visible while being learnt (during in-game tutorials and easier levels), customized (when a player changes game command keys), or blocked (because of unexpected events and/or a player's lack of mastery).

However, it is still necessary to emphasise the distinction between the player's computational text and the player's interpretative text, since these may function either independently or in close conjunction. For example, a reader of a hypertext narrative may have a mental map of the links between each page of text, but this may have no bearing on their interpretation of the text. Conversely, there may be instances when the structure of the links in a hypertext resonate with the text. Murray (1997) addresses this issue when she argues that the structure of games may complement their meaning. For example, navigating complex, closed and/or repetitive environments may give rise to “suspense, fear of abandonment, fear of lurking attackers, and fear of loss of self” (p. 135). A game, then, may produce a type of immersion and agency that complements its intended meanings or aesthetics by manipulating its procedural, participatory, spatial and encyclopaedic qualities. Murray describes the significance of several structures of navigation in this sense: story mazes, the rhizome, the journey story, problem-solving, “losing” and “winning” stories, games as symbolic dramas, the game as contest (agon), and world construction in MUDs.

The corollary of an analysis of “interactive narratives,” or, more appropriately, “ergodic texts,” is an awareness of not only the collapse between narrative and game, but also the inter-relationship between hypothesis-making about future events (cognitive-oriented interpretation of the “text”) and the performance of physical tasks (sensorimotor-oriented interaction with the “work”). In practice, narrative-oriented games require at least minimal sensorimotor action, and sensorimotor-oriented games require strategic cognition. Consequently, despite the theoretical attention it may have received, the terminological distinction between “story” and “game” is less important than the processes of engagement that give rise to the aesthetic experience of gameplay. The issue that emerges from this is how sensorimotor action and cognition are cued to work in conjunction during gameplay, and how this produces an aesthetic, emotional experience.

Emotion in Video Games: Early Models and Research

The research on emotions in video games has largely been confined to e xperimental attempts to prove the aggressive effects of gameplay upon players ( Anderson & Morrow, 1995; Ballard & Wiest, 1996; Cooper & Mackie, 1986; Dominick, 1984; Fling, Smith, Rodriguez, Thornton, Atkins, & Nixon, 1992; Funk & Buchman, 1996; Griffiths & Dancaster, 1995; Irwin & Gross, 1995; Kubey & Larson, 1990; McClure & Mears, 1984; Nelson & Carlson, 1985; Schutte, Malouf, Post-gorden & Rodasta, 1988; Scott, 1994; Silvern, 1987). The methodology of these experiments may be seen, with some variation (and qualification), as taking the following general form. First, a group of players are measured for an aggressive state. Second, some players are given a so-called “violent” game, while other players are given a so-called “non-violent” game. Third, after a measured period of play, the aggressive states of both groups are re-measured. Fourth, the subsequent difference in the level of aggression is found to be either short-lived, minimal, or non-existent.

There are many problems with this experimental methodology. Aside from Jenkins' (1990) observation that early research tended to abstract “violence” from its context of reception, the experimental conditions of these types of research produce an artificial context for producing and expressing aggressive tendencies, and tend to confuse correlation with causation (Durkin, 1995, 1999). Certainly, there are broad, historically specific correlations between the male demographic and the aesthetics associated with challenge, competition, and aggressive mastery. However, aggressive personality types often choose to engage with media that reflect and reinforce their existing tendencies, making the measurements correlative, not causal.

A more obvious criticism is that such research does not adequately account for the cathartic experience of games, that is, the extent to which emotions are aroused and purged and thereby nullified, reduced, or redirected into a socially acceptable form. Catharsis is an aesthetic evident in Malone's (1981), Myers' (1990c) and Turkle's (1984) analysis of players' preferences, but assumptions about catharsis in gameplay need qualification. First, like aggression, the catharsis players expect from games is less likely to be measurable in an unfamiliar experimental context: the everyday routines one seeks relief from are attenuated or disrupted by the experimental conditions. Second, it is simplistic to see the reduction of aggression as evidence of beneficial catharsis. To presume that aggression felt in relation to a video game is necessarily negative and anti-social, and therefore undesirable, is the same as evaluating a person's anger at their football team losing without reference to the psychological, sociological, and ideological functions of fandom. Some players may find it useful to induce manageable anger at a game rather than deal with anger, or other emotion, resulting from some unmanageable social situation.

Third, measuring a purged emotion after its purging ignores or misrepresents the desired aesthetic experience of playing a video game. D etermining whether or not some players end up with a (temporary) higher level of stress or aggression after play may be beside the point if what players seek from a game is what they experience while playing it, not the state they expect to be in afterwards. In this respect, a “black box” experimental approach draws attention away from the actual processes of gameplay, especially the real-time regulation of emotions. Video games may produce stress relief that is measurable by comparing the “before” and “after” levels of physiological arousal. However, not only may this purging be episodic, significant emotional moments of play may occur early on in, or before the end of, a gaming session, and the actual closure may be poorly defined or arbitrary. The underlying problem here is that the catharsis hypothesis was developed for traditional dramatic forms—stage plays, print novels and feature films—which have an aesthetic formula organised around discrete emotion-episodes polarised around central concerns, in which elements are drawn together and resolved in something that subsequently resembles a total, organic text. Video games have been played for the last three decades not as ineffective dramatic forms but as a different kind of dramatic form, or, more precisely, as a different kind of emotion-regulator.

Within the bulk of the research addressed in this chapter there has, it should be noted, been a tendency to consider a range of emotional states in video games. However, those emotions tends to be addressed indirectly and/or are seen as debased, compared with print and film. The alternated states of “meditation,” “kinaesthesia,” “immersion,” “agency,” and “flow” may be positive, but they are generalised from certain genres or aesthetics, principally first person shooters, and refer to general states. They therefore provide a limited basis for analysing the dynamic regulation of emotions in any particular game, and it remains necessary to provide a more systematic theoretical account of the regulation of meaning and emotion in gameplay.

Frijda's Model of Emotion

While Bordwell's (1986) account does not address the emotional experience of film viewing, researchers such as Smith (1995), Tan (1997), and Grodal (1997) see emotional responses as of equal or primary significance. For Tan (1997), while there may be many motives for watching films, “what all natural viewers of the traditional feature film have in common is their desire to experience emotion as intensely and abundantly as possible, within the safe margins of guided fantasy and a closed episode” (1997, p. 39). In an analysis of video games, in which the player's sensorimotor system is engaged and emotional responses are not merely virtual, it becomes even more necessary to appreciate the motivating role of emotion.

Following Tan, we can turn to Frijda's definition (1986, 1988, 1989) of emotions as changes in action readiness through the appraisal of a situation. “Action readiness” refers to how emotions activate the nervous system, giving way to rapid physiological changes, such as heart rate, bronchial dilation, and inhibition of the digestive system. These not only prepare an individual for physical action in a general sense, they are tied to specific “action tendencies” (Frijda, 1986, p. 70): to run, attack, seek intimacy, kiss, and so on. Some theorists, such as Izard (1984), argue that each emotion has its own neurophysiological substrate, such that a basic set of emotions can be identified. Most of these “basic” or “elementary” emotions—notably interest, happiness, surprise, anger, fear, sadness, and disgust—are generally accepted. However, cognitive (Kagan, 1984), structural-developmental (Sroufe, 1996), functionalist (Campos & Barret, 1984; Frijda, 1986), and socio-cultural (Abu-Loghod & Lutz, 1992) traditions in the study of emotion argue that the quantity and/or quality of emotions is seen as depending upon the stage of development, cognitive appraisal, and/or the cultural context (see Mascolo & Griffin, 1998, pp. 4-14).

The functionalist position is privileged here, following Frijda (1986), Lazarus (1991), and Ortony, Clore and Collins (1988), who see emotion as having a control precedence. That is, “emotion is functional in that it takes control, precluding extended reasoning: a decision to take action—any action—is better than failing to act in time” (Tan, 1997, p. 43). Emotion, then, influences one's attentiveness and motivation towards a particular problem, and, by extension, one's ability to deal with this problem. However, a key difference between theorists is the extent to which emphasis is placed upon associated physiological and cognitive processes. Without minimising the extent to which the different traditions offer conflicting interpretations, it is possible to take the moderate position that each tradition principally emphasises different aspects of processes that cross the biological and psycho-social.

Upon this basis we can appreciate Tan's (1997) distinction between “primary” and “secondary” appraisal. For Tan, following Frijda (1986), “primary appraisal” is an advanced, automated, and immediate process directly related to the stimulus, premised upon such basic elements as intensity, match-mismatch, hedonic quality, agency and causality. This evaluation is limited to “immediately observable affordances” (Tan, 1997, p. 47), that is, those forms of action which are obviously associated with the object or situation, such as fight or flight (Gibson, 1979; see also Baron, 1980). Indeed, the term “affect” can be usefully reserved for the biological component of arousal consequent upon primary appraisal (Basch, 1976). Tomkins' (1962, 1963, 1991) classic analyses of the complex assemblies of nine innate affects is paradigmatic of this type of usage. While the feeling-states that Grodal (1997) refers to are more general than Tomkins' (1962) affects, they may also be seen as affective in the limited sense that they are premised upon primary appraisal. “Secondary appraisal,” by contrast, is a continued cognitive elaboration accompanied by “emotional significance that is not directly evident from the situation itself” (Tan, 1997, p. 47) (see Frijda, 1986; Lazarus, 1966). In Frijda's (1986) terms it is only when secondary appraisal occurs that we can speak of true emotion, and this study reserves the term “emotion” for this usage. Indeed, for Frijda, each emotion has a distinctive situational meaning structure, or cognitive evaluation. For example, fear is not merely a physiological response, it is a state of arousal combined with a cognitive evaluation of an imminent threat.

A central question for the study of emotion in film is if the viewer experiences true emotions. Tan (1997), following Frijda, is aligned with the cognitive branch of film theory (Allen, 1997; Carrol, 1990; Grodal, 1998), which critiques the assumption that viewers mistake the screen for reality. Tan (1997) differentiates between an involuntary “illusion of motion,” based upon the physiological phenomena of “flicker fusion” and “apparent motion,” as well as a series of voluntary illusions: the “illusion of the diegetic effect,” “the illusion of the controlled witness,” and “the illusion of the observational attitude.” These “illusions” comprise a sequence which most viewers voluntarily and progressively accept and invest in. The viewer thereby takes up a voluntary observational position as an invisible witness and enters into the diegetic effect. Being in the position of witness the viewer knows that events do not affect his/her welfare directly, but is still able to form an emotional relationship with the film and its characters. This relationship is not one of habitation or straightforward identification, but of empathetic concern from the position of spectator. This finds its analogue in other situations in which we find ourselves spectators, such as watching sport. Even if the events being watched are not real, our relationship to situations and characters as a spectator is genuine, such that we may feel genuine, albeit inhibited, emotions. As Frijda (1989) argues: “film does not elicit emotions by products known to be imaginary, but allows inhibition of emotions because the products are known to be imaginary” (¶ 4).

Tan (1997) elaborates upon this by distinguishing between “Artefact-emotions,” or “A-emotions,” which are responses to the film's non-diegetic elements, from “Fiction-emotions,” or “F-emotions,” which are responses to situations within the film's diegesis. In the classical feature film, A-emotions are minimal (or minimised). Tan consequently focuses on F-emotions, arguing that interest is the key emotion produced while watching a feature film, and that this is oriented around the production and confirmation of expectations about events in the diegesis. These expectations are governed by “thematic structures,” such as viewers' expectations about the course of a story about “Revenge,” and “character structures,” that is, how characters' goals and motives lead viewers to privilege and expect certain outcomes on behalf of those characters (p. 179). The action tendency of interest is a desire to keep watching, but more complex emotions, such as admiration, pity, and sadness, may be reinforced by, and reinforce, interest. These more complex emotional responses are likely to be inhibited because they pertain to a fiction, and the action tendencies associated with them are virtual, or imagined. Someone watching a film may be cued to dislike a particular character on the screen, but s/he is not obliged to act upon that emotion by abandoning their seat and assaulting the screen; s/he merely imagines and fears or hopes for certain outcomes.

Frijda's (1986) model of emotion, and Tan's (1997) application of it to film, provides a basis for some valuable questions and departure points. In video games, the cognitive response to the aesthetic object gives way to some form of sensorimotor-ergodic action, such that video games produce both virtual and actual action tendencies. This being the case, how does the player's position as both invisible witness (to narrative sequences) and active participant affect his/her emotional responses? Should we see participation as giving rise to a different sort of interest, and does the confluence of virtual and action tendencies affect the course of this interest? Should we expect video game characters to elicit the same kind of empathy felt towards filmic characters? And does a player's concern with his/her own interests as a player interfere with his/her ability to empathise with characters?

Conclusion

The above issues will be addressed in the rest of this thesis. Each chapter focuses on a different theoretical concern and/or aspect of gameplay, drawing from various theories of narrative, cognition and emotion. The ongoing concern is to redress vulgar notions of “reality” and “identification” which have characterised popular opinions and early research about gameplay. The general argument in this respect is that the construction of reality may be seen in terms of aesthetics rather than ideology, and that what needs more attention is the reality of players' relationships to the games they play. These relationships cannot be reduced to assumptions about increased identification through interaction; they are better seen as aesthetic, emotional relationships which result from interest in the game and empathetic emotions directed towards characters and players from the position of witness or player.