Gamification, that is, the usage of game content in non-game contexts, has been successfully employed in several application domains to foster end users’ engagement and to induce a change in their behavior. Despite its impact potential, well-known limitations concern retaining players and sustaining over time the newly adopted behavior. This problem can be sourced from two common errors: basic game elements that are considered at design time and a one-size-fits-all strategy in generating game content. The former issue refers to the fact that most gamified applications focus only on the superficial layer of game design elements, such as points, badges and leaderboards, and do not exploit the full potential of games in terms of engagement and motivation; the latter relates to a lack of personalization, since the game content proposed to players does not take into consideration their specific abilities, skills and preferences. Taken together, these issues often lead to players’ boredom or frustration. The game element of challenges , which propose a demanding but achievable goal and rewarding completion, has empirically proved effective to keep players’ interest alive and to sustain their engagement over time. However, they require a significant effort from game designers, who must periodically conceive new challenges, align goals with the objectives of the gamification campaign, balance those goals with rewards and define assignment criteria to the player population. Our hypothesis is that we can overcome these limitations by automatically generating challenges, which are personalized to each individual player throughout the game. To this end, we have designed and implemented a fully automated system for the dynamic generation and recommendation of challenges, which are personalized and contextualized based on the preferences, history, game status and performances of each player. The proposed approach is generic and can be applied in different gamification application contexts. In this paper, we present its implementation within a large-scale and long-running open-field experiment promoting sustainable urban mobility that lasted 12 weeks and involved more than 400 active players. A comparative evaluation is performed, considering challenges that are generated and assigned fully automatically through our system versus analogous challenges developed and assigned by human game designers. The evaluation covers the acceptance of challenges by players, the impact induced on players’ behavior, as well as the efficiency in terms of rewarding cost. The evaluation results are very encouraging and suggest that procedural content generation applied to the customization of challenges has a great potential to enhance the performance of gamification applications and augment their engagement and persuasive power.

中文翻译：

---

自动生成和推荐游戏化的个性化挑战

游戏化，即在非游戏环境中使用游戏内容，已成功应用于多个应用领域，以促进最终用户的参与并引起他们行为的改变。尽管具有潜在的影响力，但众所周知的局限性在于留住玩家并随着时间的推移维持新采用的行为。这个问题可能源于两个常见的错误：设计时考虑的基本游戏元素和生成游戏内容的一刀切策略。前一个问题是指大多数游戏化应用程序只关注游戏设计元素的表面层，例如积分、徽章和排行榜，而没有充分利用游戏在参与度和动机方面的潜力；后者与缺乏个性化有关，因为向玩家推荐的游戏内容并没有考虑到他们的具体能力、技能和偏好。综上所述，这些问题往往会导致玩家感到无聊或沮丧。挑战的游戏元素提出了一个苛刻但可实现的目标和奖励完成，经验证明有效地保持玩家的兴趣并随着时间的推移维持他们的参与。然而，它们需要游戏设计师的巨大努力，他们必须定期构思新的挑战，将目标与游戏化活动的目标保持一致，平衡这些目标与奖励，并为玩家群体定义分配标准。我们的假设是，我们可以通过自动生成挑战来克服这些限制，这些挑战在整个游戏中针对每个玩家进行个性化。为此，我们设计并实施了一个全自动系统，用于动态生成和推荐挑战，这些系统根据每个玩家的偏好、历史、游戏状态和表现进行个性化和情境化。所提出的方法是通用的，可以应用于不同的游戏化应用程序上下文。在本文中，我们展示了其在一项持续 12 周并涉及 400 多名活跃参与者的大规模和长期开放式实验中的实施，以促进可持续的城市交通。进行比较评估，考虑通过我们的系统完全自动生成和分配的挑战与人类游戏设计师开发和分配的类似挑战。评估包括玩家对挑战的接受程度，对玩家行为的影响，以及奖励成本方面的效率。评估结果非常令人鼓舞，表明应用于挑战定制的程序性内容生成具有提高游戏化应用程序性能并增强其参与度和说服力的巨大潜力。