Abstract
AbstractEveryday decisions require us to predict how valuable different choice options will be in the future. Prior studies have identified a cognitive map in the hippocampal-entorhinal system that encodes relationships between states and enables prediction of future states, but does not inherently convey value during prospective decision making. Here, we investigated whether the entorhinal cortex integrates relational information about changing values by representing an abstract value space. To this end, we combined fMRI with a prospective decision making task that required participants to track and predict changing values of two choice options in a sequence. Such a sequence formed a trajectory through an underlying two-dimensional value space. Our results show that participants successfully integrated and extrapolated changes along the two value dimensions. Participants’ choice behavior was explained by a prospective reinforcement learning model and the degree to which they updated values over time correlated with self-reported navigational abilities and preferences. Crucially, while participants traversed the abstract value space, the entorhinal cortex exhibited a grid-like representation, with the phase of the hexadirectional fMRI signal (i.e., the orientation of the estimated grid) being aligned to the most informative axis through the value space. A network of brain regions, including the ventromedial prefrontal cortex (vmPFC), tracked the prospective value difference between options and the occipital-temporal cortex represented the more valuable option. These findings suggest that the entorhinal grid system might support the prediction of future values by representing a cognitive map, which might be used to generate lower-dimensional signals of the value difference between options and their identities for choices. Thus, these findings provide novel insight for our understanding of cognitive maps as a mechanism to guide prospective decision making in humans.
Publisher
Cold Spring Harbor Laboratory
Reference109 articles.
1. Abraham, A. , Pedregosa, F. , Eickenberg, M. , Gervais, P. , Mueller, A. , Kossaifi, J. , Gramfort, A. , Thirion, B. , & Varoquaux, G . (2014). Machine learning for neuroimaging with scikit-learn. Frontiers in Neuroinformatics, 8. https://doi.org/10.3389/fninf.2014.00014
2. Revealing Neurocomputational Mechanisms of Reinforcement Learning and Decision-Making With the hBayesDM Package;Computational Psychiatry,2017
3. Symmetric diffeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brain
4. Grid-like Neural Representations Support Olfactory Navigation of a Two-Dimensional Odor Space;Neuron,2019
5. The valuation system: A coordinate-based meta-analysis of BOLD fMRI experiments examining neural correlates of subjective value
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献