Double-checks or better no repeated steps? The role of effective and efficient exploration behavior for successful complex problem solving

Complex problem solving (CPS) is a prominent competence, significantly related to educational achievements. Regarding CPS, the domain-general exploration strategy VOTAT (vary-one-thing-at-a-time; Tschirgi, 1980 ) is strongly correlated with CPS success. However, the number of additional VOTAT exploration steps remained unclear. Going beyond previous studies, we combined these approaches and analyzed logfile based process data. The effectiveness (i.e., using VOTAT) and the efficiency (i.e., non-necessary, additional VOTAT-steps) of the exploration behavior were investigated across a sequence of CPS items in N = 469 high-school students. Across the item sequence, the exploration behavior became more effective and efficient. In latent class analyses, four classes were identified. Students in classes that demonstrated more effective exploration behavior reached higher intelligence and CPS performance scores than other students. Regarding the remaining classes, the relation between efficient exploration behavior with intelligence and CPS performance was less consistent. Implications for fostering students' explorations and CPS success are discussed. The results of our study emphasize the role of strategic exploration behaviors while exploring complex problem solving (CPS) tasks for successfully solving complex problems. Based on more or less effective (i.e., using VOTAT) and efficient (i.e., the number of additional, non-necessary exploration steps) strategy use, we found four different groups (i.e., latent classes) of students: (1) ineffective explorers, (2) inefficient explorers, (3) emerging explorers, and (4) proficient explorers. Students of the four classes differed in their mean intelligence and CPS scores. These results form a basis for adaptive interventions aimed at fostering beneficial exploration strategies to improve problem solving skills as an important educational goal. • We examined students while they worked on complex problem solving (CPS) items. • We focused on the effectiveness and efficiency of the exploration behavior (VOTAT). • Latent class analysis (LCA) of students' exploration behavior revealed four classes. • The classes differed in their intelligence scores and CPS performance. • The classes with effective exploration behavior showed better CPS performance.