"Oh no, the forest is burning!" cultural differences in the complex problem-solving process only under high uncertainty.

What do people in different cultures do when they encounter complex problems? Whereas some cross-cultural research exists about complex problem-solving predictors and performance, the process has rarely been studied. We presented participants from Brazil, Germany, the Philippines, and the United States with two computer-simulated dynamic problems, one where quick action was required - the WinFire simulation - and one where cautious action was required - the Coldstore simulation. Participants were asked to think aloud in their native language while working on these two tasks. These think-aloud protocols were digitally recorded, transcribed, and coded by coders in each country in terms of the steps involved in complex problem solving and dynamic decision making. For the current study, we developed a program to calculate transition frequencies from one problem solving step to another and analyzed only those protocols with more than 15 transitions. For WinFire, these were 256 think-aloud protocols from the four countries with a total of 12,542 statement, for Coldstore, these were 247 participants with a total of 15,237 statements. Based on previous, limited cross-cultural research, we predicted that after identifying a problem, Brazilians would make emotional and self-related statements, Germans would engage primarily in planning, Filipinos would gather additional information, and Americans would primarily state solutions. Results of latent transition analysis partially support these hypotheses, but only in the highly uncertain Coldstore situation and not in the more transparent WinFire situation. Transition frequencies were then also analyzed regarding community clusters using the spinglass algorithm in R, igraph. Results highlight the importance of process analyses in different tasks and show how cultural background guides people's decisions under uncertainty.