The optimal balance of controlled and spontaneous processing in insight problem solving: fMRI evidence from Chinese idiom guessing

Cognitive control is a key factor in insight generation. However, the neurocognitive mechanisms underlying the generation of insight for different cognitive control remain poorly understood. This study developed a parametric fMRI design, wherein hints for solving Chinese idiom riddles were gradually provided in a stepwise manner (from the first hint, H1, to the final hint, H4). By classifying the step-specific items solved in different hint-uncovering steps/conditions, we could identify insightful responses for different levels of spontaneous or controlled processing. At the behavioral level, the number of insightful problem solving trials reached the maximum at a intermediate level of the cognitively controlled processing and the spontaneously idea generating in H3, while the bilateral insular cortex and thalamus showed the robust engagement, implying the function of these regions in making the optimal balance between external hint processing and internal generated ideas. In addition, we identified brain areas, including the dorsolateral prefrontal cortex (dlPFC), angular gyrus (AG), dorsal anterior cingulate cortex (dACC), and precuneus (PreC), whose activities were parametrically increased with the levels of controlled (from H1 to H4) insightful processing which were increasingly produced by the sequentially revealed hints. Further representational similarity analysis (RSA) found that spontaneous processing in insight featured greater within-condition representational variabilities in widely distributed regions in the executive, salience, and default networks. Altogether, the present study provided new evidence for the relationship between the process of cognitive control and that of spontaneous idea generation in insight problem solving and demystified the function of the insula and thalamus as an interactive interface for the optimal balance of these two processes.

Increasing breadth of semantic associations with left frontopolar direct current brain stimulation: a role for individual differences

The aim of this study was to evaluate the influence of left frontopolar versus auditory (control) cortex transcranial direct current stimulation (tDCS) on the breadth of semantic associations produced in a cued free association task. A within-participants design administered anodal tDCS over the left frontopolar or auditory cortex, centered at electrode site AFZ or T7 using a 4×1 targeted stimulation montage. During stimulation, participants produced free associates in response to cues designed to promote narrow, moderate, or broad semantic associations. We measured the latent semantic associative strength of generated words relative to cues. The cue manipulation produced expected effects on the associative breadth of generated words, but there was no main effect of stimulation site, and calculated Bayes factors showed strong support for the null hypothesis. However, individual differences in creative potential, as assessed by the remote associates test, reliably and positively predicted increases in associative breadth under the frontopolar versus the auditory control condition, but only in response to narrow cues. In conclusion, the present data support neuroimaging studies demonstrating the involvement of left frontopolar cortical regions in generating relatively broad semantic associations. They also provide novel evidence that individual differences in creative potential may modulate the influence of brain stimulation on the breadth of generated semantic associations.