Theta power during encoding predicts subsequent-memory performance and default mode network deactivation.

Area of Science:

• Neuroscience
• Cognitive Psychology
• Brain Imaging

Background:

• The subsequent memory paradigm contrasts brain activity for remembered versus forgotten items to understand memory encoding.
• Previous electroencephalography/functional magnetic resonance imaging (EEG/fMRI) studies indicate an inverse link between frontal theta power and default mode network (DMN) blood oxygenation level dependent (BOLD) signals.

Purpose of the Study:

• To investigate whether the hypothesized theta-DMN relationship is less pronounced for later forgotten (LF) compared to later remembered (LR) trials.
• To elucidate the role of theta oscillations in memory encoding processes.

Main Methods:

• Utilized simultaneous EEG/fMRI in 14 healthy participants during an episodic memory task with pictorial stimuli.
• Calculated single-trial theta power (400-800 ms post-stimulus) and integrated it with fMRI data using general linear models (GLMs).
• Assessed theta's predictive power on DMN BOLD signal change via univariate GLMs and independent component analysis (ICA).

Main Results:

• Theta power was significantly higher for LR stimuli than for LF stimuli.
• The inverse relationship between theta power and DMN BOLD signal was consistently stronger for LR pictures than for LF pictures.
• EEG theta oscillations play a key role in mitigating processes that could hinder memory encoding.

Conclusions:

• Findings support the hypothesis that the theta-DMN inverse relationship is more evident for successful memory encoding (LR trials).
• Theta oscillations are integral to successful memory encoding by attenuating interference.
• This study provides further evidence for the neural mechanisms underlying memory formation and retrieval.