Artifactual power curves in forgetting

Area of Science:

• Cognitive Psychology
• Mathematical Modeling
• Memory Research

Background:

• Recent studies suggest forgetting follows a power function, challenging the traditional exponential model.
• This shift has significant theoretical implications for understanding memory decay.
• The mathematical relationship between time and forgetting is a key area of cognitive research.

Purpose of the Study:

• To investigate whether averaging exponential forgetting curves can artifactually produce a power function.
• To identify conditions under which such averaging artifacts are likely to occur.
• To provide a critical re-evaluation of studies claiming power-law forgetting.

Main Methods:

• Computational analysis of mathematical models.
• Reanalysis of previously published empirical data on forgetting.
• Simulations exploring the impact of slope variability, noise, and range restriction.

Main Results:

• Arithmetic averaging of exponential curves can indeed create an artifactual power curve.
• The artifact is pronounced when component curve slopes differ significantly.
• Artifact magnitude depends on noise distribution and performance range restrictions.

Conclusions:

• Claims of power-law forgetting should be carefully examined for potential averaging artifacts.
• The presence of range-restricted noise can also distort memory function analysis.
• Future research must account for these methodological artifacts in memory research.