Forgetting in immediate serial recall: decay, temporal distinctiveness, or interference?

Area of Science:

• Cognitive Psychology
• Memory Research
• Computational Neuroscience

Background:

• Understanding the mechanisms of forgetting from immediate memory is crucial for cognitive science.
• Existing hypotheses include time-based decay, decreasing temporal distinctiveness, and interference.
• Computational models offer a framework to test these hypotheses rigorously.

Purpose of the Study:

• To empirically test three distinct hypotheses of forgetting from immediate memory.
• To compare the predictive power of three serial recall models: primacy, SIMPLE, and SOB.
• To investigate the impact of filled delays during encoding and recall on memory performance.

Main Methods:

• Two experiments were conducted manipulating delay durations and types (articulatory suppression, choice reaction task).
• Serial recall performance was measured under various delay conditions.
• The primacy model, SIMPLE (scale-independent memory, perception, and learning), and SOB (serial order in a box) models were fitted to the experimental data.

Main Results:

• Short, filled delays (articulatory suppression) significantly impaired memory compared to no delay.
• Extending delay durations had minimal additional impact on memory recall.
• The interference-based SOB model provided the best fit to the observed data, outperforming the primacy and SIMPLE models.

Conclusions:

• Purely temporal explanations for forgetting, such as time-based decay, are insufficient.
• Interference plays a significant role in immediate memory forgetting.
• The findings support models that incorporate interference mechanisms over those relying solely on temporal factors.