• Skip to primary navigation
  • Skip to main content
  • Skip to primary sidebar
  • Skip to footer
UT Shield
The University of Texas at Austin
  • Home
  • About
  • Research
    • Safer Places
    • Soundscape-to-Landscape
    • Geographic Imagination
    • Equitable Motherhood
    • Future Cartography
  • Publications
  • People
    • Open Positions
  • Partners
  • Courses

Geographic Imagination

How can we give physical form to the places we imagine – and to the realities of place constructed in the mind?

Imagine this: you step into a snow-covered forest, where frost-laden branches bow beneath a pale winter sky. Your boots break the stillness with each crisp footfall, leaving a solitary trail through the untouched snow. Sunlight filters between the trees, scattering across the frozen ground, while your breath briefly clouds the air before disappearing.

Now picture a dimly lit underground passageway. The ground trembles as gears grind and metal bolts screech to a reluctant halt. Cigarette butts and black tar speckle the concrete as you weave through a mass of weary travelers, ticket in hand. Your ride is here.

Although neither place is named, each scene evokes a distinctive environment, atmosphere, and emotional response. What kind of place do you see? Which details shape that mental image? How does the brain transform sensory experience into a sense of place?

The Geographic Imagination research area investigates these questions by developing cutting-edge brain-decoding methods that reconstruct perceived environments from neural activity. Through brain-to-image generation and quantitative evaluation, we examine not only whether a viewed place can be reconstructed, but also whether the resulting image preserves its geographic character and perceptual meaning.

Current work centers on Mind2Place, a multimodal brain-to-image framework that brings together EEG, fMRI, and MEG through neural encoding and diffusion-based image generation. The model reconstructs places viewed by participants, offering a new way to study how environments are represented across different forms of brain activity.

We are also further developing Place Similarity Score (PSS) metrics (as seen in Soundscape-to-Landscape) to evaluate Mind2Place and other contemporary brain-decoding models. These metrics assess how closely reconstructed images resemble the places that participants originally viewed—not only in visual appearance, but also in their spatial, geographic, and perceptual qualities.

Together, this work seeks to deepen our understanding of how people perceive, imagine, and mentally represent the places with which they interact.

Primary Sidebar

Recent Posts

  • GISense Lab at AAG 2026
  • Undergrad Andy Qin Presented at ACM SIGSPATIAL 2025
  • GISense Lab at ACSP 2025 & Visit to the University of Minnesota
  • Visiting Scholar Koichi Ito at the GISense Lab
  • Dr. Hongxu Ma at DeepMind Visited our Lab

Footer

GISense Lab

We explore human-centered GeoAI, GIScience, and spatial data science, to understand human experience at place in a responsible way.

Quick Links

About

People

Courses

Contact

2201 Speedway, WCP 4.160

Austin, TX 78712

gisenselab@outlook.com

 

  • Email
  • GitHub
  • LinkedIn
  • Twitter

UT Home | Emergency Information | Site Policies | Web Accessibility | Web Privacy | Adobe Reader

© The University of Texas at Austin 2026