Playing Nice

At the other "Theory/Social Impact panel" at ACE 2008, which I chaired, the emphasis was either on the use of interactive technologies to simulate realistic conversations and interpersonal social exchanges or on integrating better games for children with the public welfare agendas of the state. However, because play is often subversive, in many ways it could be argued that the former goal is less problematic than the latter, although Joseph Weizenbaum -- the creator of ELIZA -- famously argued that encouraging people to converse with computers could also have corrosive social effects.

Chris Swain started the session with a talk about ELECT BiLAT, a military training game for improving bilateral negotiations in the Iraq war that I mention briefly in the forthcoming Virtualpolitik book and discuss at considerable length in an essay in the forthcoming Joystick Soldiers collection. (In writing the article, I interviewed Swain and three other designers who worked on the game and discussed how competing interface design philosophies shaped how trust was represented in the game.) Swain's presentation on "The Augmented Conversation Engine – A System for Achieving Believable Conversation in Games and Interactive Stories" argued that the game was designed to teach "soft skills" in the theatre of war to foster "culturally sensitive and better negotiation."

Like many teachers of game design, Swain began by defending his medium with a familiar argument by looking to the primitive aesthetic and dramatic products of early silent cinema. In particular, he emphasized those from the decade that saw the technical novelty of Edward Muybridge's 1884 photographic sequences evolve into the more intimate close-up of "Fred Ott's Sneeze." He argued that imitation of a prior medium is a strong tendency in all new media and that, just as early filmmakers reproduced the representative techniques of plays, videogames would become more realistic as a result of creative and not technical breakthroughs. If film ultimately arguably became "the literature of the twentieth century," Swain argued that a 1960 game of Spacewar makes it difficult to imagine videogames as "the literature of the 21st century."

In explaining the rationale of his team's ELECT BiLAT game and what Swain saw as its larger significance, he emphasized the fact that physical choice plays a large role in videogames, while verbal choice plays a very small one, which is almost the exact inverse of the situation of real people making significant conversational choices constantly as part of their daily lives. Conversation is also obviously something that is hard to simulate, a fact that I can attest to since I carpool to work with a specialist in artificial intelligence who sometimes observes that our everyday conversational moves and the inferences deployed in the language we use in the carpool lane would be impossible for a contemporary computer to simulate.

Swain argued that more realistic player-to-NPC dialogue could prove to be as revolutionary for videogames as "adding sound to movies." Currently, he complained that too often conversations were too obviously "on rails" in the branching tree of that came down to little more than "which porridge Goldilocks should taste." To illustrate his point about the nuances of spoken discourse, he showed stills of a mobster from the Sopranos both giving information and lying to an FBI agent. Unfortunately, the alternative of sophisticated AI-based programs were susceptible to language tie-ups, Swain claimed. In demonstrating the conversation engine in the game he showed how it modeled a "sea of possible actions," which could be particularly important for a "high-context culture" like Iraq. Swain showed a whole range of conversational gambits with just one character that included "Tell GIs to dig hole," "Ask about the curse, and "Talk about soccer." (His approach, which required seven writers, was apparently critiqued by computer scientists, who objected that little computational ingenuity was required for this brute force discursive approach.)

To continue the discussion about the potential for designing compelling conversational intelligent agents, next up was Stanford's Steven Dow who designed the locative graveyard installation "Voices of Oakland" when he was a graduate student at Georgia Tech and was presenting at ACE about his experiences playtesting the 3-D augmented reality version of the interactive drama Façade, which was originally created for the keyboard interface by Michael Mateas and Andrew Stern. (When Dow's installation was part of the Grand Text Auto exhibit at the Beall Center, I took my digital rhetoric class for a tour of the works in the gallery.) Based on 33 participants, Dow argued that he had isolated five specific styles of play in AR Façade, which were somewhat different from Bartle's classic four player types, and included the following stock characters: Engager, Performer, Tinkerer, Observer and Partaker. To demonstrate the relevance of his typology, he showed video of users in the installation. In Dow's talk, I was most interested to find out that the system wasn't run by natural language input, as one would assume, because there was actually a Wizard of Oz behind the scenes who input the player's phrases and gestures.

Perhaps the youngest presenter at the conference was MIT junior Eletha Flores who talked about games for stroke rehabilitation, a subject that I've talked about some here before, in connection with the LIRT system. Flores argued that groups like the Elder Games Project were conducting research on how to make the repetitive elements of stroke rehabilitation more enjoyable. However, Flores argued that many elderly people preferred games like Trivial Pursuit to whack-a-mole/whack-a-mouse pastimes. She also reviewed the ARMin System to argue that her collaborators at fatronik tecnalia may be pursuing more user-centered design.

Given the moral panics about videogames in the United States, it was particularly interesting to see the presentation of Jordi Sánchez-Navarro about the role of non-formal education taking place in Esplais, which is what state-sanctioned leisure associations for children and young people are called in Spain, where the contribution of ideas, creativity, imagination, helping, participation, and cooperation are valued as social objectives by local government. Sánchez-Navarro argued that local officials and even the PTA understood that traditional playing and gaming would need to be updated for digital society and culture. He also pointed to Henry Jenkins's 8 myths about videogames to legitimate their efforts. Of course, critics familiar with the work of the researchers associated with the Grand Theft Childhood study might question if it was really "free choice" for the students in the group if M-games were prohibited.

Finally Tilde Bekker argued that technology could enhanced traditional play in ways that invited more imagination than conventional videogames. In a talk illustrated with a juice box with a straw, she argued that similar devices for interactions such as rolling, shaking, and tilting could provide feedback about behavior that had intuitive appeal for young people. She showed images of children playing with very simple devices, such as red/green/blue game parts with changing colors and black boxes with embedded simple technologies for skate parks.

The closing keynote was provided by Ryuta Kawashima who presented research associated with the popular serious game Brain Age 2, which is aimed to mitigate the depressing fact that as we increase in semantic competence during our lifetimes, we face a linear cognitive decline after age twenty. (You can see how Kawashima is a character in the game, by looking at the website for the Japanese version.)

The game is a rare success in the serious games movement, one which has been both lucrative and well-reviewed, with the exception of this interesting critique by Ian Bogost. Like many Internet celebrities, Kawashima has treated himself as an experimental subject and showed pictures of his own brain as he engaged in complex tasks such as riding a motorcycle. He noted, however, that the critical prefrontal cortex could also be stimulated by mundane tasks such as reading aloud, handwriting, and simple arithmetic that light up many parts of brain. He also presented research intended to make more nuanced arguments about recreational videogames than Akio Mori's negative image of videogame-playing brains atrophying in a state of "game brain" with studies on both fighting games and role-playing games.