Background. Differet types of creative thinking such as divergent and convergent tasks supposedly have reciprocal relationships with affective experiences of people participating in tasks. Previous studies have affirmed that affects influence on creative performance. Recent studies have revealed that creative thinking influence people's affect. Divergent thinking appears to increase the positive mood of participants, whereas convergent thinking appears to displays the opposite. However, a creative project often involves both divergent and convergent tasks interlace in an order. Combinations and orders of divergent and convergent tasks are supposed to provoke affects in intertwined ways. This study aimed to investigate how the order of divergent and convergent tasks in creative process influence change in affective experiences of group work participants. We hypothesized that change in affects would not only be influenced by the types of tasks but also by the predecessor process.

Methods. Twenty-two graduate students taking the graduate course "Design Thinking" in an engineering school for one semester participated as subjects. A series of questionnaire-based investigations were conducted in four class meeting which had different combinations of divergent and/or convergent tasks. Each class meeting involved two tasks. In the first class students performed two convergent tasks in a series. The second class started with a divergent task followed by a convergent task, while the third class started with a convergent task followed by a divergent task. Finally, in the fourth class two divergent tasks were performed in a series. The subjects were measured by their affective states in terms of valence and arousal with the self-assessment manikin scale three times in each class; before the first task, between the first and second tasks, and after the second task. The valence and arousal levels were compared between three measurement times to derived change in affects. Changes in affects were compared with a series of paired t-test between types of tasks: divergent and convergent, as well as between the first and second tasks.

Results. Valence and arousal of subjets were scattered around the neutral level in the first measurement time in almost all classes. Generally, after conducting the convergent process, emotions remained around the neutral level. However, emotions shifted to more positive after divergent tasks. After two consecutive convergent tasks in the first class, affect score changed to more negative. Meanwhile, the other three class meeting showed the similar patterns where affect score changed to more positive and aroused. A series of paired t-tests showed that valence significantly increase in all divergent tasks. The decrease of valence was observed only in the convergent-convergent combination in the first class. Likewise, arousal level significantly increased after almost all divergent tasks, while the decrease of arousal level observed in convergent-convergent combination in the first class was not statistically significants. The changes in affects was also analyzed as the Euclidian distance in valence-arousal plane. A series of paired t-tests with the distance as an index of affect change proved that the change in affects was more significant in the first process rather than in the second process except for convergent-divergent combination in the third class. Additionally, the distance of the score after the first process of divergent-convergent combination during the second class was significantly larger than the distance after the first process of convergent-divergent combination of the third day.

Conclusions. The change in participants affective states raised more significantly in the divergent process rather than through the convergent process. The convergent task did not necessarily turn down valence and arousal. Only in the case of the class with the full convergent process, the affect change to negative. The current study revealed how strong divergent process induced emotion. Moreover, the first task plays a significant role to leverage participants' affective experiences.

Added-Values. The current study illustrated the importance of the order and combination of tasks in creative group works. Particularly this study emphasized the importance of the first process in multiple tasks. This study implied how to design creative tasks in terms of optimal affective experiences of participants.

Keywords: affect change; creative process; divergent thinking; convergent thinking