Spatial Cognition

Spatial cognition research investigates how humans and animals represent and navigate space, focusing on understanding the underlying neural mechanisms and computational processes. Current research explores the parallels between human spatial reasoning and the capabilities (and limitations) of artificial intelligence models, particularly in vision-language models and 3D reconstruction systems, often employing transformer-based architectures and algorithms inspired by hippocampal place cells and grid cells. These studies reveal critical gaps in AI's understanding of spatial relationships and perspective-taking, highlighting the need for improved model architectures and training methodologies, while also offering valuable insights into human cognitive disorders like constructive apraxia. Ultimately, advancements in this field will improve AI systems' ability to interact with and understand the physical world, and enhance our understanding of the neural basis of human spatial cognition.