Schedule of Events | Search Abstracts | Invited Symposia | Symposia | Rising Stars | Poster Sessions | Data Blitz

Medial temporal default mode network selectively encodes autobiographical visual imagery

Poster Session A - Saturday, March 7, 3:00 – 5:00 pm, Fairview/Kitsilano Ballrooms

andrew anderson¹ (andanderson@mcw.edu); ¹Medical College of Wisconsin

The human brain’s capacity to imagine visual scenes from memory is thought to rely on the medial temporal subsystem of the default mode network, yet the neural codes supporting this ability remain poorly understood. We combined functional magnetic resonance imaging (fMRI) with artificial intelligence models to characterize these codes during autobiographical imagination. Fifty participants imagined re-experiencing twenty natural scenarios while undergoing fMRI, when cued by generic text prompts (e.g., party, funeral, driving, exercising). Individual scenes were modeled using Stable Diffusion to generate personalized images from verbal descriptions of the scenarios imagined, collected before participants underwent fMRI. These depictions were then transformed into image-recognition network embeddings. Representational Similarity Analysis revealed that the medial temporal default mode network encoded participant-specific representational structure of visual embeddings, even when controlling for semantic features derived from a large language model. This effect was absent in other networks and during reading without imagination, identifying the MT-DMN as a core substrate for self-generated visual scene reconstruction.

Topic Area: LONG-TERM MEMORY: Other