The induction of a specific mental set for problem solving is accompanied by increased neural representational similarities

Abstract

A mental set is a cognitive bias induced by the successive application of a specific strategy; developing a specific mental set prevents individuals from generating more direct and creative ways of solving novel problems. The neural mechanisms underlying mental set induction are unclear. We predicted that the repeated application of a specific strategy leads to increasing similarity of the related neural representations. To test this hypothesis, representation similarity analysis (RSA) was used to identify brain regions whose neural encoding patterns showed increases in representational similarity (RS) during the first, second, third, and fourth induction trials, which utilized the same type of ordinary radical-level “loose” chunk decomposition (LCD) strategy. This induction would consequently restrict one’s ability to solve the probe trials that required a different strategy (insightful stroke-level “tight” chunk decomposition, TCD). We found that from the first to the final LCD trial, regions for attentional control, mental operation, and the default mode network (DMN)-based processing showed increased RS. These RS increases positively predicted the activation of the anterior cingulate cortex (ACC) and medial frontal lobe when solving TCD probe trials, which in turn modulated the impacts of the RS increase in the process of mental set inducing on the degree of representational changes in regions for perceptual, operative, and executive processing on TCD trials. Thus, increased similarities of neural representational patterns in the goal-directed manipulation and spontaneous processing DMN systems may underlie mental set induction; these neural changes interacted with conflict resolution and representational changes when solving novel problems.