A hippocampal–parietal network for learning an ordered sequence

The dynamics of the neural network that underlies learning transitive structures of an ordered sequence remains poorly understood. To address this, in the present study we used fMRI to track the time course of transitive inference learning. The hippocampus and the angular gyrus were each shown to be closely related to the learning trajectory, but differentially so. Hippocampal activity was shown to consistently increase with learning but no correlation was found between performance and hippocampal activation, suggesting a general role for the hippocampus. Left angular gyrus activity was also found to consistently increase with training, but, in addition, correlated significantly with behavioral performance. This suggests an involvement of the angular gyrus in learning the ordinal associations between the stimuli.