Evidence Against Novelty-Gated Encoding in Serial Recall.

Area of Science:

• Cognitive Psychology
• Neuroscience
• Memory Research

Background:

• Novelty-gated encoding posits that novel events are more strongly encoded into memory than familiar ones.
• This is a fundamental assumption within the SOB model of serial recall.
• The current study investigates the validity of this assumption through empirical testing.

Purpose of the Study:

• To empirically test predictions derived from the novelty-gated encoding assumption in serial recall.
• To determine if memory encoding is influenced by the novelty of preceding items.
• To explore alternative explanations if novelty-gated encoding is not supported.

Main Methods:

• Three experiments were conducted to assess recall probability under varying conditions.
• Experiment 1 manipulated phonological similarity of preceding items.
• Experiments 2 and 3 varied item classes and distractor types within a complex-span paradigm.

Main Results:

• Recall probability was not dependent on phonological similarity of preceding items (Experiment 1).
• Memory performance was unaffected by whether preceding items belonged to the same or different classes (Experiment 2).
• Phonologically similar distractors did not impair memory more than dissimilar ones (Experiment 3), contradicting predictions.

Conclusions:

• The findings question the core assumption of novelty-gated encoding in serial recall.
• Results suggest that novelty may not be the primary driver for encoding strength in this context.
• The study highlights potential boundary conditions for the role of prediction error in memory acquisition and points to alternative explanations for previous findings.