Space Wars episode 4: A new HOPI
(higher-order path integration)
Space Wars episode 4: A new HOPI (higher-order path integration)
- Aaron Blaisdell -
University of California, Los Angeles, USA
Animals that must return to previously visited locations in space, such as a home nest, food and water sources, etc., depend on spatial learning. Theoretical accounts of spatial learning fall into two camps, bottom-up behavioral processes and top-down cognitive processes. I review the history of the debate on spatial learning – the Space Wars in four episodes. Episode I begins with the rat maze learning experiments of Small (1901) and the early debate between behaviorist S-R and mentalistic accounts of spatial learning. Episode II recounts Hull’s neobehaviorist account in terms of fractional goal responses. In Episode III, I cover O’Keefe and Nadel’s groundbreaking work on the hippocampus as a cognitive map. Finally, in Episode IV, I discuss our new model of spatial learning and inference that involves higher-order path integration (HOPI). HOPI combines features of each approach, behavioral and cognitive. I end by discussing empirical support for the HOPI model, and how it calls into question Tolman’s cognitive and O’Keefe and Nadel’s neural representational accounts of spatial inference.