The Role of Curiosity1 and Curiosity2 in the Emergence of Insight

We propose that the processes underlying insight problem solving in humans depend on two distinctly different forms of curiosity: Curiosity1 (which is associated with a habit-based, goal-centered, reinforcement learning processing system), and Curiosity2 (which depends on the discursive, default mode, medial-temporal-lobe based processing system). The former kind of curiosity is goal directed and increases with approach to the rewarding answer. The latter is exploratory and goal averse: “twiddling.” The possibility of insight, we suggest, depends upon the individual initiating a deliberate system switch upon apprehension of an impasse. Problem solving involves engaging in a habitual mode of responding and motivation by Curiosity1. With insight problems, however, this normal mode fails to lead to a solution, and impasse results. Acknowledgment of the impasse may trigger a strategic switch to a different kind of curiosity and information processing system: the discursive, default mode Curiosity2 system, wherein the solution that was previously unavailable may be found. This view is consistent with traditional stages posited to be involved in insight problem solving. However, several paradigms used to study creativity or investigate 'Aha!' reactions do not fit easily with this view of insight. Using this perspective, we evaluate the evidence for insight in nonhuman animals.