Navigation representations during active navigation are predominantly goal-directed
Tianjiao Zhang, Jack Gallant, University of California, Berkeley, United States
Session:
Posters 1 Poster
Location:
Pacific Ballroom H-O
Presentation Time:
Thu, 25 Aug, 19:30 - 21:30 Pacific Time (UTC -8)
Abstract:
Previous experiments revealed that the human brain represents many different navigational features. However, because most fMRI experiments study individual representations in isolation, the relative importance of these many different features to navigation remains unclear. To compare these representations, we recorded BOLD activity while subjects performed a taxi driver task in a large, realistic virtual world. Voxelwise modelling was performed with 21,283 stimulus- and task-related features that encompass 33 different types of information that might be represented during naturalistic navigation. The fit models show that navigational information is represented broadly across the cortex, including in many areas outside known navigation-related ROIs. Among navigational models, goal-directed representations account the most variance, while passive perceptual representations account for much less variance. These data suggest that representations during active naturalistic navigation are predominantly goal-directed.