Thesis

Modern video games have reached a new level of graphic fidelity. Furthermore with the rise of popularity of new technologies such as Virtual Reality and Augmented Reality, gamers are more immersed than ever within virtual worlds and virtual characters. The credibility and believability of Virtual Characters (typically called NPCs) requires them to have basic human traits such as emotions and the ability to make decisions on their own. One of the most defining human attributes is our social ability and awareness. Academic research groups have created numerous different social architectures that can transform video game characters and change their impact on the game’s world. Characters have social desires, complex behaviours and work towards changing the social status around them. However, academic research on AI in games and commercial game AI development efforts are rather disjoint and there is a wide divergence in methods and results. This is the result of academic researchers wanting general solutions to generic problems whereas video game developers want something that works well enough. This project’s goal is to implement a social architecture model, originated from Academic Research, in a modern and commercially successful video game and investigate its feasibility and impact on player experience. We present and describe an implementation of CIF-CK: an agent social architecture: “Comme il Faut” in a commercial videogame “The Elder Scrolls V: Skyrim”. The implementation was distributed as a downloadable Mod, and met with almost instant success in the player community.

With the mobile games market growing by the day, more and more gamers are shifting to mobile platforms. Most of the games, however, are very simplistic, which leads to non interesting gameplay for veteran players, or are ports of preexisting games, which leads to poor interfaces and user experience. This project focuses on the second problem, trying to understand what the best user interface/user experience is in tactical turn based games for mobile platforms.

Believable interactions between synthetic characters are an important factor defining the success of a game relying on the player being able to create emotional bonds with the game characters. As important as the character being themselves believable is that the interaction with or between such characters is believable. Although research in synthetic characters has developed several models to improve character believability, interactions are generally not the focus of such works. This may be one of the reasons why state of the art models from Academia are still not being used in commercial products. In this thesis, we bridged affective computing and traditional animation principles and create a model for character interaction based on anticipation and emotion that allows for precise affective communication of intention-based behaviors. We also present a study with 52 subjects supporting that our proposal is able to increase scene believability when compared to traditional approaches.

The motivation for this work came from the belief that our civilization is becoming more connected to the ”virtual” world, and that users are using virtual communication tools to be able to communicate, mostly without the need to communicate face to face. This leads to a more closed environment, where people exit their houses only when needed (get groceries, go to work, etc). By viewing positive effects of games on players, and using them in non-game contexts, gamification was created, and is already being used and researched with many approaches, but there is no research regarding its use on outdoor activities, as a motivator for people to exit their homes. For this purpose, we approached this topic and developed GeoChest as a possible solution. GeoChest combines Gamification and GeoCaching, an outdoor activity, in an attempt to understand which gamification techniques and game mechanics can bring a positive feedback from the users that use the application. In the end, we concluded that not only the application was well-designed, but also that most of the application’s game mechanics implemented were well-received by the users, ending this dissertation with the belief that we were able to bring some interesting contributions to Gamification research.