Dynamic changes in semantic memory structure support successful problem-solving

Creative problem-solving is central in daily life, yet its underlying mechanisms remain elusive. Restructuring (i.e., reorganization of problem-related representations) is considered one problem-solving mechanism and may lead to an abstract problem-related representation facilitating the solving of analogous problems. However, empirical evidence supporting such mechanisms is scarce. We used network science methodology to estimate participants’ individual semantic memory networks (SemNet) before and after attempting to solve a riddle. These networks represent the organization of solution-relevant and -irrelevant concepts as nodes, with edges representing the strength of the relationship between them based on participants’ relatedness judgments. Restructuring was quantified as the difference in SemNet metrics between pre- and post-solving phases. We found that restructuring a problem-related SemNet was associated with the successful solving of this problem and an analogous one. We showed that both specific solution-relevant concepts and concepts that were semantically remote became more strongly related in solvers. We also showed that only changes in semantically remote concepts were instrumental in actively solving the riddle while changes in solution-relevant concepts probably reflect a pre-exposure to the solution. These results shed new light on mental restructuring associated with problem-solving and analogical transfer, and show how changes in SemNet can capture this restructuring.

Auteurs Bordeaux Neurocampus