Building Up Serious Games with an Artificial Life Approach: Two Case Studies

Artificial Life (AL) studies how to reproduce life-like phenomena exploring the life as could be in artificial systems (software, hardware, or hybrid). This challenging scientific perspective has produced a number of programming techniques often applied to solve concrete problems (Data Analysis, Process Optimization, Social Simulations, etc.). Computer Gaming is a field where AL techniques are applied. There are many successfully Alife products for pure entertainment (e.g., Tamagotchi and Creatures) and for educational objectives (e.g., Avida-Ed). However, we notice that all AL-Based games share a general flavor: they refer in someway to biological scenarios. In other terms, they represent often a sort of popularization of AL experiments designed for non-scientists. In this paper we argue that AL programming techniques (or more basically bio-inspired computational algorithms) could be used to develop generic games (e.g., sports, adventures, business games, etc.) without any relation with a biological perspective. We describe BreedBot and Learn2Lead, two Serious Games that we think could be paradigmatic examples about how to use AL techniques in different ways and fields that could be very different from their biological roots. BreedBot and its sequels (BestBot and BrianFarm) have been developed to disseminate the core-concepts of Autonomous Robotics and Learn2Lead has been developed to teach Psychological Theories of Teamwork in Small and Medium Enterprises. In BreedBot, AL techniques are used to develop the player–game interaction and they are explicitly visible by the user (he/she has to train/evolve a population of artificial agents). At the opposite side, Lear2lead has an old style appearance but it hides an AL engine. In this case AL techniques are used to model the game mechanics (e.g., artificial team dynamics and avatars’ behavior). Both games are also able to be played online (www.nac.unina.it/bestbot2 and www.unina.l2l.it).