The development of numerical estimation: evidence for multiple representations of numerical quantity

Area of Science:

• Cognitive Psychology
• Developmental Psychology
• Mathematics Education

Background:

• Prominent models of numerical representation (logarithmic-ruler and accumulator models) do not fully explain children's estimation abilities.
• Understanding how numerical representations develop is crucial for addressing learning difficulties.

Purpose of the Study:

• To investigate children's and adults' numerical estimation and the underlying representations.
• To evaluate existing models of numerical representation against empirical data.
• To identify factors influencing the choice and development of numerical representations.

Main Methods:

• Participants (second graders, fourth graders, sixth graders, and adults) completed two numerical estimation tasks.
• Tasks were performed in two distinct numerical contexts to assess representational flexibility.
• Analysis focused on the consistency of estimation patterns with different numerical representation models.

Main Results:

• Data contradicted both the logarithmic-ruler and accumulator models for all age groups.
• Children demonstrated the use of multiple numerical representations, with increasing reliance on appropriate ones with age and experience.
• The choice of numerical representation was influenced by the specific numerical context.

Conclusions:

• Children's numerical estimation difficulties may stem from a reliance on logarithmic representations when linear representations are required.
• Numerical representation is not static but develops dynamically with age, experience, and context.
• Current models need revision to account for the flexibility and context-dependency of numerical representations in children and adults.