Map making: Constructing, combining, and inferring on abstract cognitive maps

Abstract

Cognitive maps are thought to enable model-based inferences from limited experience that can guide novel decisions-a hallmark of goal-directed behavior. We tested whether the hippocampus (HC), entorhinal cortex (EC), and ventromedial prefrontal cortex (vmPFC)/medial orbitofrontal cortex (mOFC) organize abstract and discrete relational information into a cognitive map to guide novel choices. Subjects learned the status of people in two separate unseen 2-D social hierarchies defined by competence and popularity piecemeal from binary comparisons. Although only one dimension was ever behaviorally relevant, multivariate activity patterns in HC, EC and mOFC were linearly related to the Euclidian distance between people in the mentally reconstructed 2-D space. Hubs created unique comparisons between the two hierarchies, enabling inferences between novel pairs of people. We found that both behavior and neural activity in EC and vmPFC/mOFC reflected the Euclidian distance to the retrieved hub, which was reinstated in HC. These findings reveal how abstract and discrete relational structures are represented, combined, and navigated in the human brain.Kriete, T., Noelle, D. C., Cohen, J. D., & O'Reilly, R. C. (2013). Indirection and symbol-like processing in the prefrontal cortex and basal ganglia.

Modeling awake hippocampal reactivations with model-based bidirectional search

“…The model further predicts that these experimental results are not simply reflecting memory consolidation of things that have been previously well learned, but are more generally related to the learning process itself. Consistently, the hippocampus has been found to be necessary for learning in several reward-based navigation tasks (Girardeau et al, 2009;Jadhav et al, 2012;de Lavilléon et al, 2015) and interactions between the hippocampus and the orbitofrontal cortex have been recently proposed to contribute to exploit the task structure learned by the internal model to infer the value of stimuli within the environment (Jones et al, 2012;Klein-Flügge et al, 2013;Wikenheiser and Schoenbaum, 2016;Zhou et al, 2019;Park et al, 2019).…”

Protocol for building a cognitive map of structural knowledge in humans by integrating abstract relationships from separate experiences

“… PsychoPy code to run the training protocols Park et al., 2020a Open Science Framework ( https://osf.io/bnc3w/ ) with the identifier https://doi.org/10.17605/OSF.IO/bnc3w . MATLAB code for behavioral analysis and simulation Park et al., 2020a Open Science Framework ( https://osf.io/bnc3w/ ) with the identifier https://doi.org/10.17605/OSF.IO/bnc3w . MATLAB code for the fMRI data analyses Park et al., 2020a https://github.com/argmaxv/socialmap …”

“… MATLAB code for behavioral analysis and simulation Park et al., 2020a Open Science Framework ( https://osf.io/bnc3w/ ) with the identifier https://doi.org/10.17605/OSF.IO/bnc3w . MATLAB code for the fMRI data analyses Park et al., 2020a https://github.com/argmaxv/socialmap …”

“…Here, we introduce a behavioral training protocol designed for human participants to implicitly build a map of two-dimensional social hierarchies while making a series of binary choices and analytic tools for measuring the internal representation of this structural knowledge. For complete details on the use and execution of this protocol, please refer to Park et al. (2020a , 2020b) .…”

“…For complete details on the use and execution of this protocol, please refer to Park et al. (2020a , 2020b) .…”