Abstract
This article inspects theory emerging from literature review and labora-tory work on games for cultivating creative literacy. Whether games can or cannot instil or alienate one’s creativity is debatable. On one hand, they can be risk-and stress-free exploring grounds for people to interact in ways without parallel; on another, they can disengage players from the real world. Nevertheless, they have the potential to be turned into tools for thinking, for learning and for articulating knowledge between individu-als. With that into consideration, we pinpointed two main groups that branch into four major categories: Behaviours – comprised of Attitudes and Competencies – and Conditions – comprised of Procedures and Resources – which we are structuring into a framework from which we draw hypotheses that undergo validation through play-testing sessions, in order to improve the framework.

Abstract
Background: Blended treatments, combining digital components with face-to-face (FTF) therapy, are starting to find their way into mental health care. Knowledge on how blended treatments should be set up is, however, still limited. To further explore and optimize blended treatment protocols, it is important to obtain a full picture of what actually happens during treatments when applied in routine mental health care. Objective: The aims of this study were to gain insight into the usage of the different components of a blended cognitive behavioral therapy (bCBT) for depression and reflect on actual engagement as compared with intended application, compare bCBT usage between primary and specialized care, and explore different usage patterns. Methods: Data used were collected from participants of the European Comparative Effectiveness Research on Internet-Based Depression Treatment project, a European multisite randomized controlled trial comparing bCBT with regular care for depression. Patients were recruited in primary and specialized routine mental health care settings between February 2015 and December 2017. Analyses were performed on the group of participants allocated to the bCBT condition who made use of the Moodbuster platform and for whom data from all blended components were available (n=200). Included patients were from Germany, Poland, the Netherlands, and France; 64.5% (129/200) were female and the average age was 42 years (range 18-74 years). Results: Overall, there was a large variability in the usage of the blended treatment. A clear distinction between care settings was observed, with longer treatment duration and more FTF sessions in specialized care and a more active and intensive usage of the Web-based component by the patients in primary care. Of the patients who started the bCBT, 89.5% (179/200) also continued with this treatment format. Treatment preference, educational level, and the number of comorbid disorders were associated with bCBT engagement. Conclusions: Blended treatments can be applied to a group of patients being treated for depression in routine mental health care. Rather than striving for an optimal blend, a more personalized blended care approach seems to be the most suitable. The next step is to gain more insight into the clinical and cost-effectiveness of blended treatments and to further facilitate uptake in routine mental health care.

Abstract
The structure and evolution of a scientific research community can be quantitatively assessed taking into account the interactions between scientific agents dispersed geographically. In the recent years, CSCW has stabilized as a cross-disciplinary field suffering significant changes in its core structure, and there is limited understanding about the factors influencing the nature and progress of collaborative computing research. In this paper, we measure the correlation between a set of features related to the influence of collaboration types on the number of citations as well as the geographical distribution of the accumulated contribution to the CSCW literature. Overall, our work can represent a starting point to demonstrate how the study of scientific collaboration can partly explain the variations in the number of citations, frequency of papers, and topics addressed.

Abstract
Much research attention on crowd work is paid to the development of solutions for enhancing microtask crowdsourcing settings. Although decomposing difficult problems into microtasks is appropriate for many situations, several problems are non-decomposable and require high levels of coordination among crowd workers. In this chapter, we aim to gain a better understanding of the macrotask crowdsourcing problem and the integration of crowd-AI mechanisms for solving complex tasks distributed across expert crowds and machines. We also explore some design implications of macrotask crowdsourcing systems taking into account their scaling abilities to support complex work in science.

Abstract
Crowdsourcing has shown to be a valuable problem-solving approach to handle the increasing complexity and scale of tasks for which the current AI algorithms are still struggling. Crowd intelligence can be particularly useful to train and supervise AI systems in a symbiotic, co-evolutionary relationship that raises long-term research challenges to the hybrid, crowd-computing design space. With the increase in the scale of mixed-initiative approaches, we need to gain a better understanding of the implications of crowd-powered systems as a scaffold for AI through the study of massive crowd-machine interactions. In this paper, we identify some open challenges and design implications for future crowd-AI hybrid systems. A framework is also proposed based on the practical challenges of addressing human-centered AI methods and processes.

Abstract
A substantial amount of work is often overlooked due to the exponential rate of growth in global scientific output across all disciplines. Current approaches for addressing this issue are usually limited in scope and often restrict the possibility of obtaining multidisciplinary views in practice. To tackle this problem, researchers can now leverage an ecosystem of citizens, volunteers and crowd workers to perform complex tasks that are either difficult for humans and machines to solve alone. Motivated by the idea that human crowds and computer algorithms have complementary strengths, we present an approach where the machine will learn from crowd behavior in an iterative way. This approach is embodied in the architecture of SciCrowd, a crowd-powered human-machine hybrid system designed to improve the analysis and processing of large amounts of publication records. To validate the proposal's feasibility, a prototype was developed and an initial evaluation was conducted to measure its robustness and reliability. We conclude this paper with a set of implications for design.