Abstract:  The insight problem solving has been proposed to involve different phases: First, there is an initial structuring phase in which the problem’s starting conditions, goal and possible actions, as given in the problem statement, are represented internally. A second general stage involves searches within the problem space, which may take the form of means-ends analysis or forward search. However, the initial structuring phase is misleading and solution cannot be reached without re-structuring of the problem representation. Thus, with insight tasks, re-structuring of the problem representation is necessary for the problem to be solved. However, the debate of the psychological mechanism focused on the representation restructuring process, and there are two interpretations: consciousness processing and unconscious processing. We conducted three studies to explore the psychological mechanism of the insight problem solving. Study 1 showed there is a significant positive correlation between the executive functions and the insight problem solving, and updating of working memory representations can predict individual insight scores. Study 2 investigated verbal insight problems, using behaviors and ERPs to explore the effect of executive functions on insight problems. The results showed that the insight, compared with no-insight impasses, elicited a more positive potential P2 and in the time windows 620~800 ms (P3). P2 may reflect that people perceive intuitively mental impasses at the perceptual stage, whereas the P3 is associated with the process of constraint relaxation characterized by the alternation between the old and new ways of thinking. These results suggest that executive function influences searching within the problem space, but not the re-structuring of the problem representation.

Key words:  顿悟, 表征重组, 执行功能, 无意识, 事件相关电位