Dr. Yangwen Xu | Disentangling cognitive maps and graphs in the human brain

Cognitive maps and graphs are two fundamental forms of mental representations structuring knowledge. A cognitive map localizes identities in a Euclidean space, but a cognitive graph defines identities among relations. Whether the neural foundations of these two mental representations are dissociated remains a mystery. Previous studies grounded their hypotheses on one representation without considering the other, which leads to potential confounding between map- and graph-like representations. This fMRI study aims to disentangle these two representations in the human brain by letting participants learn a Euclidean graph, where the representational contents of maps and graphs are orthogonalized. We will use univariate contrasts, adaption analyses, and representational similarity analyses to localize the neural representations of cognitive maps and graphs. This study will shed light on the long-lasting problem of whether cognitive maps and graphs rely on different neural foundations.