Purchase this article with an account.
Liwei Sun, Sebastian Frank, Peter Tse; A Graph-like Neural Representation of Indoor Spaces Revealed Using fMRI. Journal of Vision 2018;18(10):744. doi: 10.1167/18.10.744.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
An internal spatial representation of the indoor structure of a familiar building guides us to our destinations within the building. Here, we investigated how interconnectedness between subspaces (rooms) in a building is represented in the brain. We designed a square-shaped space with four corner rooms connected by four equal-length corridors in a virtual reality world. Two of the corridors were partitioned into three subspaces (rooms), while the other two were non-partitioned. Participants were trained to memorize the indoor space while searching for target objects fixed in the corner rooms as landmarks. Participants were fMRI scanned while viewing the target objects before and after the training. Results reveal a significant increase in multivariate classification accuracies of object-room associations in the occipital place area (OPA) after training. Moreover, the objects separated by the partitioned corridor were represented as further away than the objects separated by the non-partitioned corridor in the parahippocampal place area (PPA) after training, measured by neural distance (1 - correlation). This result suggests that the human brain codes the indoor spaces in a manner analogous to a graph-like representation with subspaces (rooms) as nodes.
Meeting abstract presented at VSS 2018
This PDF is available to Subscribers Only