Contributed Talk Sessions | Poster Sessions | All Posters | Search Papers

Poster Session C: Friday, August 15, 2:00 – 5:00 pm, de Brug & E‑Hall

Modeling the Visualization of Personal Experiences during Imagination in Medial Temporal DMN

andrew james anderson¹; ¹Medical College of Wisconsin

Presenter: andrew james anderson

Imagination enables the human brain to relive personal experiences by mentally simulating visual scenes. This ability has been linked to the Medial Temporal subsystem of the brain’s Default Mode Network (MT-DMN). However, the representational codes associated with visualizing personal experiences have been understudied, in part because quantitatively modeling freeform imagination is challenging. To target this, we scanned fifty peoples’ brain activity with fMRI as they reimagined their personal experience of twenty diverse natural scenarios (e.g. wedding/funeral/driving). To model the visualization of personal experiences, we deployed image-generation AI models to depict participants’ verbal self-reports of their mental images (made outside the scanner) and image recognition models to re-represent the synthetic image features in a more abstract visual form, invariant to view and scale. A Representational Similarity Analysis suggested that MT-DMN selectively reflected visual model structure, when controlling for semantic features derived from language models. This effect was not observed in other brain networks which were more sensitive to the language model. This finding helps characterize the neural bases of imagination, and earmarks image AI models as valuable tools for neurally decoding imagination.

Topic Area: Brain Networks & Neural Dynamics

Extended Abstract: Full Text PDF