An integrative model of network activity during episodic memory retrieval and a meta-analysis of fMRI studies on source memory retrieval.

Functional magnetic resonance imaging studies on episodic memory (EM) retrieval have used a variety of paradigms/contrasts to yield many diverse findings, which raises the need for an integrative multifactorial view. To address this need, I begin by proposing a fourstage, quasi-sequential model of network activity during episodic remembering, termed "the Retrieval, Experience, and Decision (RED) model." The model combines EM-retrieval macro-functions (retrieval effort, retrieval, subjective experience of remembering, and decision effort), macro-structures (frontoparietal control network, default mode network, and medial temporal lobe) and essential experimental factors (encoding strength, retrieval specificity, retrieval confidence, and pretrial expectation) into a broad, unified frame of reference. Next, to evaluate the validity of the model, I formulate four multifactorial predictions aimed at three distinct types of source-memory contrast (source-hit > source-miss, source-hit > item-hit, and source-hit > correct-rejection), and test them using a meta-analytic method. The results of the meta-analyses are largely consistent with the predictions, supporting the RED. In conclusion, the RED provides an overarching framework that configures coherently with many representative findings in the literature, thereby offering a useful foundation capable of guiding further research efforts on related topics.