In this project we aim to distinguish sequential expectations from non-sequential expectations that are driven by a cued episodic retrieval.
The basis for prediction is memory. From this perspective, memory is not autotelic, but should be optimized to serve the anticipation of upcoming events and the planning of action. This optimization entails updating when the world has truly changed. In the current project, we will trigger the recall of and modify episodic memories either with regard to sequential expectations (based on the episodic memory trace) or with regard to non-sequential expectations (based on semantic information). In a first step, participants will be videotaped while performing and observing everyday actions. Subsequently, three experiments will be conducted using BOLD-sensitive functional Magnetic Resonance Imaging to assess the cerebral basis of episodic expectation, surprise (information-theoretical: surprisal), and re-consolidation during presentation of original and modified action videos. Using this episode modification paradigm (Schiffer et al., 2012, 2013), we employ a set of novel experimental factors concerning the episodes’ mnemonic solidity (retrieval times and consolidation) and experiential quality (self-perspective and self-performance) to test their impact on sequential and object-semantic surprise. This approach is motivated by the question as to which conditions render the memory of a truly experienced episode more or less susceptible to later modification of its spatiotemporal structure or its object-semantic content. Moreover, we systematically compare the conditions for the presence of memory updating effects due to reconsolidation separately for sequential structure and object-semantic content. Behavioral analyses will be combined with BOLD fMRI contrast, representational similarity, and graph theoretical analyses to specifically determine the role of hippocampal and selected cortical areas in stable and flexible episodic memory.