Modeling distributions of immediate memory effects: no strategies needed?

Area of Science:

• Cognitive Psychology
• Memory Research

Background:

• Existing immediate memory models lack validation for predicting effect size distributions.
• Understanding effect size distributions is crucial for accurately modeling memory phenomena.

Purpose of the Study:

• To evaluate if the feature model predicts appropriate effect size distributions for phonological confusion and word-length effects.
• To assess the model's ability to account for performance variations and reversals in immediate memory tasks.

Main Methods:

• Utilized the feature model (Nairne, 1990) to analyze immediate memory data.
• Examined the model's predictions against observed distributions of effect sizes, including reversals.

Main Results:

• The feature model successfully predicted appropriate effect size distributions for phonological confusion and word-length effects.
• The model accurately accounted for the number of reversals observed in participants' performance.
• It correctly predicted that lower overall performance correlates with smaller, less reliable effects and increased reversals.

Conclusions:

• The feature model offers a robust explanation for immediate memory effects without invoking changes in encoding, rehearsal, or storage buffers.
• The distribution modeling approach shows promise for broader applications in memory research.