Interactive problem solving has been proposed as an experimental manipulation that significantly increases the success of solving various matchstick algebra problems by allowing solvers to interact with physical representations of the problems. In contrast to this claim, we hypothesized that the influence of interactivity would vary based on the specific sources of difficulty inherent in the problems: perceptual chunks and cognitive constraints. We carried out a conceptual extended replication across three experimental series with conditions of varying degrees of interactivity, but failed to reproduce interactive solutions amongst our participants. A followup analysis of motor activity showed that the movements of the solvers did not contribute to chunk decomposition but significantly interfered with the relaxation of higher-level constraints. These findings suggest that motor activity can hinder performance when it does not align with the cognitive demands of the task. We therefore call for a more targeted and problem-specific understanding of how physical interaction contributes to restructuring in insight problem solving. Keywords cognitive constraints, insight problem solving, interactivity, motor activity, perceptual chunks
Спиридонов et al. (Thu,) studied this question.