On the assessment of procedural knowledge: From problem spaces to knowledge spaces

Area of Science:

• Cognitive Science
• Theoretical Computer Science
• Educational Psychology

Background:

• Extends Stefanutti and Albert's (2003) work on problem-solving.
• Integrates concepts from Newell and Simon's (1972) problem-solving framework.
• Builds upon Doignon and Falmagne's (1985, 1999) and Falmagne et al.'s (2011, 2013) research on knowledge spaces and learning spaces.

Purpose of the Study:

• To provide mathematical foundations for connecting problem-solving and knowledge assessment.
• To establish that the set of all knowledge states for a problem space constitutes a learning space.
• To develop and illustrate an algorithm for deriving learning spaces from problem spaces.

Main Methods:

• Generalization and completion of prior theoretical work.
• Development of a novel algorithm to derive learning spaces from problem spaces.
• Application of the algorithm to the Tower of London (TOL) task.

Main Results:

• Demonstrates that a problem space mathematically defines a learning space.
• Successfully derives the learning space for the Tower of London (TOL) task.
• The derived learning space enables adaptive assessment of planning skills.

Conclusions:

• The theoretical framework provides a robust bridge between problem-solving and knowledge assessment.
• The developed algorithm offers a practical method for constructing learning spaces.
• The application to the TOL task highlights the potential for adaptive assessment in neuropsychological testing.