Neural Global Pattern Similarity Underlies True and False Memories

Area of Science:

• Cognitive Neuroscience
• Neuroimaging
• Computational Neuroscience

Background:

• Neural processes underlying human memory strength are not fully understood.
• Formal computational models suggest global matching plays a key role in memory strength.
• Existing research lacks a comprehensive mechanistic explanation for memory strength signals.

Purpose of the Study:

• To test the hypothesis that memory strength, for both true and false memories, is determined by the global similarity between neural activation patterns during retrieval and encoding.
• To investigate the distinct contributions of different brain regions to memory strength signals.
• To explore the relationship between neural global pattern similarity and frontal monitoring processes.

Main Methods:

• Functional magnetic resonance imaging (fMRI) to measure brain activity.
• Computational modeling to formalize hypotheses about memory strength.
• Analysis of encoding-retrieval neural global pattern similarity (ER-nGPS) in various brain regions.
• Correlation analysis between ER-nGPS and recognition memory performance.

Main Results:

• Multiple ER-nGPS signals were identified, carrying distinct information about memory strength.
• ER-nGPS in parietal regions reflected semantic similarity and scaled with recognition strength for true and false memories.
• ER-nGPS in the visual cortex uniquely contributed to true memory strength.
• Differences in ER-nGPS between parietal and visual cortices correlated with frontal monitoring processes.

Conclusions:

• Memory strength signals arise from global neural pattern similarity, with distinct contributions from different brain regions.
• Parietal ER-nGPS accounts for memory strength of both true and false memories, while visual cortex ER-nGPS is specific to true memories.
• These findings provide a mechanistic understanding of memory strength in recognition by integrating computational and neuroimaging approaches.