Decoding memory-guided predictions in the medial temporal lobe and visual cortex
Poster Session C - Sunday, March 30, 2025, 5:00 – 7:00 pm EDT, Back Bay Ballroom/Republic Ballroom
Dingrong Guo1 (guo@psych.uni-frankfurt.de), Javier Ortiz-Tudela2, Yee Lee Shing1,3; 1Department of Psychology, Goethe University Frankfurt, 2Mind, Brain and Behaviour Research Center (CIMCYC); Department of Experimental Psychology, University of Granada, 3Brain Imaging Center, Goethe University Frankfurt
How we perceive the visual environment relies on the integration between concurrent contextual information and our existing memories or prior knowledge. The predictive processing framework suggests that this integration is supported by the hierarchical and/or laminar organizations of the visual cortex, as well as its interactions with the medial temporal lobe (MTL), especially the hippocampus and entorhinal cortex (EC). To investigate the neural basis of this integration, we collected ultra-high-field 7T fMRI data using an occluder paradigm designed to isolate signals from mnemonic information and signals from concurrent contextual cues. In this 2-day experiment, participants first learned cartoon images of real-world locations featuring key objects. Twenty-four hours later, while inside the scanner, they were asked to mentally retrieve the missing objects that were located in the occluded part of the learned images. Planned data analyses will include layer-specific classifiers and representational similarity analysis to delineate different types of feedback signals across cortical layers of our regions of interest (i.e., early visual cortex, object-selective cortex and MTL subregions). Additionally, connectivity analyses will explore the interactions between these regions and their layer-specific connectivity patterns. The findings of this study are expected to enhance our understanding of how predictions are formed through the dynamic interplay between sensory inputs and pre-existing memory representations.
Topic Area: LONG-TERM MEMORY: Episodic