Investigating user-centered human computation games : the roles of perceived enjoyment and its antecedents

Abstract

The popularity of games has motivated their adoption for pursuits beyond entertainment. An emerging strategy is the Human Computation Game (HCG), the phenomenon of harnessing human intelligence through enjoyable gameplay to address computational problems that are beyond the power of computer programs but trivial for humans. As such, HCGs should be enjoyable for individuals so that their intention to play is enhanced, thereby increasing the generation of useful outputs. With the increased use of HCGs in various human computation contexts, different genres of HCG have emerged. Yet, there is a dearth of research examining the multidimensional aspect of players’ perceived enjoyment and its antecedents across HCG genres.

Driven by these research gaps, two main research questions are identified: 1) what factors drive the perceived enjoyment of HCGs, and 2) how the influences of these driving factors differ across HCG genres. To address the research questions, the entire research is divided into two interrelated studies. In Study I, the first research question is addressed. Based on a review of the extant literature on both entertainment- and task-oriented contexts, this study proposed a conceptual model, which suggests that personality traits and perceived output quality are influential factors driving the perceived enjoyment of HCGs. Utilizing a custom-developed mobile content-sharing HCG named SPLASH, the proposed conceptual model was tested empirically by analyzing self-reported survey data gathered from 205 participants. The results demonstrated that perceived output relevancy had a stronger influence on perceived affective enjoyment, followed by personality traits of extraversion and openness, as well as perceived output accuracy. Furthermore, perceived output relevancy had the strongest impact on perceived cognitive enjoyment, followed by perceived timeliness of output and the neuroticism trait. The variables found to predict behavioral enjoyment were perceived relevancy and the openness trait. In Study II, the second research question is addressed. Three variants of mobile content-sharing applications were developed: a collaborative HCG (Collabo), a competitive HCG (Clash), and a non-gaming application (Share). A mixed-design experiment using a total of 160 participants was conducted, with the application types as a within-subjects factor and personality traits as a between-subjects factor. Participants used all three applications on an Android-based mobile phone, and they then completed questionnaires on the respective applications. This study established that the collaborative HCG was effective in promoting the perceived enjoyment of players who scored high on the traits of extraversion, agreeableness, and openness, whereas the competitive HCG was effective for those who scored low on these traits. Moreover, individuals perceived that the output of the competitive HCG was more accurate than that of the collaborative HCG. Most importantly, the output generated by the non-gaming application was perceived to have a higher level of accuracy, completeness and relevancy compared to both HCG genres. By combining Study I and II, this research highlights the joint contribution of personality traits and game genres to explaining players’ perceived enjoyment in HCGs. In addition, this research provides evidence about the effects of HCG genres on perceived output quality. Therefore, game designers should place high priority on creating personality-targeted HCG designs, and develop quality control mechanisms or game features that could enhance players’ perceptions of output quality.