Connectome-based predictive modeling: A new approach of predicting individual critical thinking ability

Abstract

Listen

Translate article

Critical thinking (i.e., CT), which is a metacognitive process, encompasses seven dimensions: truth-seeking, open-mindedness, analyticity, systematicity, self-confidence, inquisitiveness, and maturity. It enables individuals to arrive at reasonable conclusions and make well-informed decisions. Despite being regarded as a fundamental survival skill and a crucial element of higher education, the neural mechanism responsible for a highly critical brain remains unclear.To address this, we recruited 80 Chinese native speakers, half of whom majored in liberal arts and half in science. They underwent a resting-state fMRI scan and completed a classic CT scale. Using connectome-based predictive modeling, we identified brain connectivity patterns associated with high and low CT dispositions. Then a support vector machine model was trained to classify individuals’ gender and major from the top 20 edges respectively for high- and low CT networks.Results showed that the high CT network included limbic, temporal, parietal, and cerebellum lobes. The limbic and parietal lobes of high CT networks positively predicted overall and each dimensions of CT scores. Specifically, the connectivity between the left parahippocampal gyrus and right precuneus cortex significantly predicted the open-mindedness score, the connectivity between the bilateral hippocampal gyrus significantly predicted the systematicity score, and the connectivities between the right precuneus cortex and left superior parietal lobule, right middle temporal gyrus and right superior parietal lobule were found to significantly predict individual cognitive maturity. And the top 20 edges of high CT networks could distinguish participants' major but not gender, while the top 20 edges of low CT networks could not distinguish either.These above results suggest that CT dispositions can be forecasted from a person's distinctive mapping of brain connections, with the limbic lobe of the high CT network predicting open-mindedness and systematicity scores, which share common neural mechanisms with creative thinking, and the parietal lobe positively predicting self-reported cognitive maturity.