IntroductionThis study looked at how an assessment instruction and test anxiety (TA) can influence divergent thinking (DT) components of creativity in a playful robotic problem-solving task.MethodsWe measured TA and creative performance (TD) under assessment and non-assessment conditions in 122 secondary students engaged in creative problem solving (CPS).ResultThe aspects of DT (fluidity and originality) showed a tendency to be impacted by assessment instruction. Thus, under non-assessment conditions, the learners show higher fluidity and better originality in the first occurrence of the CPS task. In the second occurrence, time spent on CPS decreases. Moreover, the originality turns to be impaired in the second trial and only student under assessment maintain their engagement in the activity. No correlation was found between TA and DT, and no gender or age differences were observed. The results suggest that TA does not influence the performance of the students involved in creative problem-solving processes.DiscussionWe discuss the findings in relation to game-based learning specificities. The assessment instruction in playful activities can be perceived as a positive challenge and even the students showing higher levels of test anxiety do not perceive it as a threat. Furthermore, if time constraints are minimized, the impact of assessment instruction on creative performance might be further reduced. The finding of this study opens promising perspectives to the research on innovative forms of school assessment and creative problem solving.