Default mode network connectivity predicts individual differences in long-term forgetting: Evidence for storage...

Area of Science:

• Neuroscience
• Cognitive Psychology
• Computational Neuroscience

Background:

• Declarative memory maintenance and forgetting mechanisms remain poorly understood.
• Measuring memory fading rates and their neural correlates is empirically challenging.
• Existing research lacks ground-truth measures for memory decay.

Purpose of the Study:

• To investigate the neural correlates of individual differences in memory forgetting speed.
• To link functional brain connectivity patterns with behavioral measures of long-term memory (LTM) forgetting.
• To explore the role of the default mode network (DMN) in memory maintenance.

Main Methods:

• Computational phenotyping to estimate individual forgetting speeds from accuracy and response time data.
• Utilized an adaptive paired-associate learning task for 33 participants.
• Applied machine learning to resting-state fMRI connectivity data, correlating it with individual forgetting speeds.

Main Results:

• Individual forgetting speeds in LTM were significantly associated with resting-state functional connectivity.
• Connectivity patterns within the default mode network (DMN) and between the DMN and sensory areas predicted forgetting speed.
• High prediction accuracy (r = 0.77) was achieved using connectivity patterns alone.

Conclusions:

• Resting-state connectivity, particularly within and connected to the DMN, plays a crucial role in memory maintenance.
• Findings support the storage degradation hypothesis of forgetting.
• The DMN and associated sensory regions are actively involved in preventing memory decline.