Cognitive processes in well-defined and ill-defined problem solving

Authors

  • Gregory Schraw,

    Corresponding author
    1. Department of Educational Psychology, 1313 Seaton Hall, The University of Nebraska-Lincoln, USA, Lincoln, Nebraska 68588
    • Department of Educational Psychology, 1313 Seaton Hall, The University of Nebraska-Lincoln, Lincoln, Nebraska, 68588, USA
    Search for more papers by this author
  • Michael E. Dunkle,

    1. Department of Educational Psychology, 1313 Seaton Hall, The University of Nebraska-Lincoln, USA, Lincoln, Nebraska 68588
    Search for more papers by this author
  • Lisa D. Bendixen

    1. Department of Educational Psychology, 1313 Seaton Hall, The University of Nebraska-Lincoln, USA, Lincoln, Nebraska 68588
    Search for more papers by this author

Abstract

We investigated the relationship between two kinds of problem solving using Kitchener's model of hierarchical cognitive processing. We predicted that performance on well-defined problems (i. e. those with a single, guaranteed solution) would be independent of ill-defined problems (i. e. those with multiple, non-guaranteed solutions). We also predicted that self-reported epistemic beliefs (i. e. assumptions about the nature and acquisition of knowledge) would be related to ill-defined, but not well-defined, solutions. Results confirmed these predictions. We concluded that well-defined and ill-defined problems require separate cognitive processes and that epistemic beliefs play an important role in ill-defined problem solving. These findings supported Kitchener's three-level model of problem solving.

Ancillary