Sources of interference in item and associative recognition memory

Area of Science:

• Cognitive Psychology
• Neuroscience
• Computational Modeling

Background:

• The global matching framework explains recognition memory by assessing cue similarity to memory contents.
• Retrieval contributions are categorized as matches and mismatches (item noise, context noise, background noise).

Purpose of the Study:

• To present a model that quantifies interference contributions in recognition memory.
• To estimate the magnitude of item noise, context noise, and background noise.

Main Methods:

• Developed a model to parameterize matches and mismatches to item and context cues.
• Applied a hierarchical Bayesian framework to 10 recognition memory datasets.
• Manipulated variables like strength, list length, and stimulus class.

Main Results:

• Item noise contribution was small and stimulus-dependent, minimal for single words/scenes.
• Background noise dominated retrieval across most stimulus classes.
• High-frequency words showed equal background and context noise.

Conclusions:

• The majority of recognition memory interference originates from pre-episodic experiences (background noise).
• This challenges models focusing solely on intra-list interference.
• Findings highlight the significant impact of prior knowledge on memory retrieval.