This paper presents research conducted by the authors to explore creativity using online large-scale data and network science methodology. We will first explain the studies in which large-scale data are used to analyze creative activities in online communities or what ideas of others led to enhancing creativity. In addition, we will introduce a study showing that large-scale data can be used as complementary data to behavioral experimental data as well as to establish metrics to evaluate creativity. The implications of such creativity research are discussed.

Recently, researchers have proposed a more valid process for performing and scoring divergent thinking (DT) tasks and new approaches to automatically score DT tasks using natural language processing and machine learning. Despite the recent proposals for this approach, its usefulness has been gaining attention. However, few comprehensive reviews have been conducted on the measurement advances in DT tasks in Japan. Therefore, this study summarizes issues related to scoring validity from the implementation of DT tasks and then conducts a systematic review focusing on the automated scoring approach for DT tasks. The current study examined (1) the types of participants, (2) the types of DT tasks, and (3) the types of automated scoring methods used for DT tasks. The results indicated that the automated scoring of DT tasks was conducted primarily on English responses to the language versions of alternative-use tasks, primarily for college students and adults in general. The most common automated scoring method is semantic distance based-scoring. However, scoring methods using supervised learning and large-scale language models have been proposed recently. Future research should examine the feasibility of using automated scoring methods by scoring DT tasks with more diverse participant populations and responses from speakers of more varied languages.

Divergent thinking (DT) has long been studied as a part of creative potentials; however, the extensive scoring process for DT poses issues. A recent solution gaining attention is automated scoring by calculating semantic distance. The divergent association task (DAT), a simplified version of the existing DT task, allows participants to engage in DT tasks within a shorter time, with immediate and automatic scoring. This study developed and validated the Japanese version of the DAT (DAT-J). Twelve candidate word vectors derived from the Japanese corpus were prepared to calculate DAT scores, and three surveys were conducted to examine the validity and reliability of the DAT-J. In Survey 1, 95 art majors completed an online questionnaire, including the DAT-J and creativity-related scales. Because Survey 1 revealed that the DAT-J may include inappropriate responses for automatic scoring, we modified the DAT-J instructions and developed a web application to validate inappropriate responses. In Survey 2, 358 students participated in the modified version of the DAT-J, the existing DT tasks, and the creativity-related questionnaire. The DAT-J was retested two weeks later. In Survey 3, an experiment was conducted comparing DAT-J responses by handwritten, web app, and web app with validation. Survey results demonstrate the basic validity and reliability of the DAT-J, underscoring its potential utility for assessing DT. Furthermore, the DAT-J answering format did not affect DAT scores. Finally, limitations of the DAT-J and future improvements for measurement accuracy were discussed.

Creativity comprises two cognitive components: convergent and divergent. Although the neural basis of divergent thinking has been well-studied, evidence for convergent thinking is limited. To our knowledge, only a few studies have been conducted on the neural basis of each type of thinking using the same participants. Therefore, the present study was designed to test the hypothesis that convergent and divergent thinking are supported by different neural mechanisms. During functional magnetic resonance imaging (fMRI), participants were asked to perform tasks of creativity to induce convergent and divergent thinking. The Remote Association Task (RAT) and Alternative Uses Task (AUT) were used to measure convergent and divergent thinking, respectively. Neuroimaging results revealed that the insular and frontal cortices were activated during the convergent thinking task, suggesting that convergent thinking is a process with a high cognitive load and primarily requires executive functions. In contrast, in the divergent thinking task, the left hemisphere, which is the language-dominant hemisphere, was mainly activated, including the left precuneus/posterior cingulate cortex, left parahippocampal gyrus, and left angular gyrus. Given that these regions are associated with autobiographical and episodic memories, it is likely that divergent thinking requires language-based long-term memory retrieval. These results suggest that the neural bases of convergent and divergent thinking differ, supporting the hypothesis that each type of thinking is a distinct component of creativity. The present findings represent an important step toward systematically understanding human creativity and exploring ways to enhance it.

This study aims to elucidate the impact and nature of the actability of tangible objects on creative thinking. We investigated how the ability to interact with physical objects influences creativity evaluation and the creative utilization of these objects through experimentation. Our hypothesis posits that the degree of actability affects the creative use of objects and the evaluation of created works. The experiment involved three conditions of actability: (1) tangible objects within reach without obstructions, (2) tangible objects out of reach without obstructions, and (3) tangible objects within reach but obstructed by a case. We compared the impression evaluations when participants viewed images of the same objects to those under the aforementioned conditions. Additionally, we administered an idea generation task and compared the number of ideas generated across conditions. The results demonstrated that the condition with tangible objects within reach without obstructions yielded a higher number of ideas and better evaluations compared to the other conditions. These findings suggest that actability influences creative thinking and provides insights into the presentation methods of artworks in contexts such as exhibitions.

Proficiency in higher-order learning, i.e., meta learning, is required in situations where discontinuous environmental changes is the norm. This study explores factors affecting meta learning in managers from this perspective, using an insight problem-solving model in cognitive science. A seven-year field study of a general manager in an IT service company revealed that, as suggested by the model of insight problem-solving, two factors promote meta learning: increased fluctuation (increased variety of trials) and better vision (increased appropriateness of evaluation), each accompanied with emergent goal setting through improvisational interaction with others and artefacts. By grasping meta learning in this way, the possibility was identified of viewing higher-order expertise in the workplace, which has been viewed as sudden, discontinuous changes, as continuous ones.

While some artists feel threatened by AI that generates works of art such as novels, paintings, and musical compositions, others see great potential in using generative AI as a tool to perpetually present works that embody their unique authorship. This paper aims to provide concrete examples of how generative AI stimulates the creativity of experts and the kinds of narratives it can evoke. The analysis focuses on interviews with a commercial cartoonist and a digital artist, respectively, in which they shared their impressions, differences from their authorship, and potential applications when confronted with the image generation AI trained on their own works and the illustrations it generated. By categorizing the statements from these interviews, two categories emerged: “authorship as qualitative calculations” and “authorship as a manifestation of ties to generative AI.” The study demonstrates that organizing a field in which artists engage with generative AI based on different criteria and calculations fosters a dialogue that relativizes and reexamines authorship from multiple perspectives. Additionally, it reveals that in the collaboration between the artifact and the artist the perception of authorship is transformed, as authorship is creatively expressed and they inspire each other.

To investigate the “lyric-first” approach to lyric writing in Japanese popular music, this study seeks to elucidate the lyric writing process by analyzing a chapter entitled “The Documentary of Lyric Writing” in the book by renowned lyricist Rei Nakanishi, in which he details his lyric writing techniques (Nakanishi, 2020). Nearly half of the process involved the lyricist’s self-evaluation and reflection on ideas, demonstrating the transformation of personal concepts into lyrics through a bottom-up process in the context of the Japanese language. Based on this analysis, an observational study was conducted with two additional lyricists to determine whether the same mechanism is applicable to others. The results show that one lyricist, unfamiliar with the lyric-first method, produced divergent lyrics in separate sections without a convergence of the monitored ideas. The other lyricist showed similarities to Nakanishi but used a different strategy by incorporating active visualization early in the writing process. These three cases suggest that the lyric-first method, particularly in the context of Japanese lyrics, is a highly specialized literary technique that requires simultaneous attention to word-level details and the overall structure of the lyric.

The arts exist within social and cultural contexts, influencing activities such as learning, creating, and teaching. Prior research on artists’ expertise has often focused on individual processes, leaving the intergenerational succession of knowledge—specifically how expert artists mentor younger generations—relatively unexplored. In this study, we interviewed two composition faculty members teaching creative composition at a music college to examine their own learning experiences and their current teaching practices at conservatories. The findings indicate that composition learning resources can be categorized into materials (e.g., scores, books, and recordings), instructors, peers, and opportunities to hear compositions. Teaching composition was found to involve the coordination and management of these resources. The teachers encouraged students to engage with a broad range of musical works and acquire a variety of knowledge about art. Furthermore, by having students organize and manage performances where their own work is showcased, they are encouraged to acquire practical knowledge not attainable through lessons alone, such as listening to the performances of their own work. During lessons, the teachers valued the students’ creative intentions and originality, encouraging them to articulate their ideas and addressing practical technical problems such as instrumentation and notation. However, they avoided imposing strong directives on what needed correction. This approach reflected an attitude of “active non-inheritance,” where the teachers refrained from having students simply follow their own compositional methods or aesthetic interests.

This review paper proposes to include the process of discovering new perspectives as part of the sequence of creativity evaluation processes. First, it reviews the existing knowledge on creativity evaluation, pointing out that creativity evaluation of products has traditionally been discussed as a static process that convergently determines whether the necessary requirements for creativity are met based on specific evaluation criteria. Next, it discusses evidence that the perspective discovery process, which involves the reconstruction of perspectives, plays a role in creativity evaluation, highlighting the importance of viewing creativity evaluation as a dynamically changing process. Finally, the paper suggests that understanding this perspective discovery process as a form of creative problem solving could lead to a more refined explanation of the creativity evaluation process and the development of effective intervention methods for evaluation.