Abstract: Creative thinking, which refers to the process by which individuals produce a unique, valuable product based on existing knowledge, experience, and multi-perspective thinking activities, is the cornerstone of human civilization and social progress. As an important part of the creative field, scientific inventions in particular require individuals to break the existing state and build new things in the process of creating them. Therefore, the use of real-life examples of scientific inventions to explore the cognitive neural mechanism of creative thinking has become a focus of recent research. There have been many studies of creative problem solving, especially regarding its neural mechanisms. However, less attention has been paid to the issue of problem finding. Hence, the present study employed resting-state functional magnetic resonance imaging (rs-fMRI) and scientific invention problem-finding materials to identify the neural substrates of the process of scientific innovation problem finding.
In the present study, nine scientific innovation problem situations were selected as materials. Each problem consisted of three parts: (paradoxical) problem situation, (misleading) old problem, and heuristic prototype. The modified learning-testing paradigm was used to explore the brain mechanisms of problem finding. Participants were asked to find a new problem based on the given problem situation and old problem in the testing phase after learning all the heuristic prototypes in the learning phase. A total of 104 undergraduates (mean age = 19.26 ± 0.99) were enrolled in the final experiment. The rs-fMRI data were acquired using an echo planar imaging (EPI) sequence from a 3-T Siemens Magnetom Trio scanner (Siemens Medical, Erlangen, Germany) at the MRI center of Southwest University. We used both the amplitude of low-frequency fluctuation (ALFF) and resting-state functional connectivity (RSFC) to measure the local properties of rs-fMRI signals, and then investigated the relationship between ALFF/RSFC and individual differences in scientific problem finding.
After controlling for age and sex, the results of multiple regression analysis showed that individuals with a high rate of useful problems had higher spontaneous brain activity in the left medial prefrontal cortex (L-mPFC) and cerebellum. Functional connectivity analysis further found a significant positive correlation between the rate of useful problems and the mPFC-Cuneus functional connectivity.
Based on these results, we infer that: (1) The mPFC plays an important role in the process of scientific innovation problem finding. It might be responsive to two aspects: one involved in breaking the thinking set and forming novel association and another associated with the extraction and processing of working memory. (2) The cerebellum and the cuneus might be separately involved in the inter-semantic allocation of attentional resources and divulging.

Key words: creativity, scientific innovation problem finding, mPFC, AlFF, RSFC