SummaryBackgroundDespite decades of research on spatial memory, we know surprisingly little about how the brain guides navigation to goals. While some models argue that vectors are represented for navigational guidance, other models postulate that the future path is computed. Although the hippocampal formation has been implicated in processing spatial goal information, it remains unclear whether this region processes path- or vector-related information.ResultsWe report neuroimaging data collected from subjects navigating London’s Soho district; these data reveal that both the path distance and the Euclidean distance to the goal are encoded by the medial temporal lobe during navigation. While activity in the posterior hippocampus was sensitive to the distance along the path, activity in the entorhinal cortex was correlated with the Euclidean distance component of a vector to the goal. During travel periods, posterior hippocampal activity increased as the path to the goal became longer, but at decision points, activity in this region increased as the path to the goal became closer and more direct. Importantly, sensitivity to the distance was abolished in these brain areas when travel was guided by external cues.ConclusionsThe results indicate that the hippocampal formation contains representations of both the Euclidean distance and the path distance to goals during navigation. These findings argue that the hippocampal formation houses a flexible guidance system that changes how it represents distance to the goal depending on the fluctuating demands of navigation.

Cognition and behavior emerge from brain network interactions, such that investigating causal interactions should be central to the study of brain function. Approaches that characterize statistical associations among neural time series-functional connectivity (FC) methods-are likely a good starting point for estimating brain network interactions. Yet only a subset of FC methods ("effective connectivity") are explicitly designed to infer causal interactions from statistical associations. Here we incorporate best practices from diverse areas of FC research to illustrate how FC methods can be refined to improve inferences about neural mechanisms, with properties of causal neural interactions as a common ontology to facilitate cumulative progress across FC approaches. We further demonstrate how the most common FC measures (correlation and coherence) reduce the set of likely causal models, facilitating causal inferences despite major limitations. Alternative FC measures are suggested to immediately start improving causal inferences beyond these common FC measures.

Most neuroscientific studies have focused on task-evoked activations (activity amplitudes at specific brain locations), providing limited insight into the functional relationships between separate brain locations. Task-state functional connectivity (FC) -statistical association between brain activity time series during task performance -moves beyond task-evoked activations by quantifying functional interactions during tasks. However, many task-state FC studies do not remove the first-order effect of task-evoked activations prior to estimating task-state FC. It has been argued that this results in the ambiguous inference "likely active or interacting during the task", rather than the intended inference "likely interacting during the task". Utilizing a neural mass computational model, we verified that task-evoked activations substantially and inappropriately inflate task-state FC estimates, especially in functional MRI (fMRI) data. Various methods attempting to address this problem have been developed, yet the efficacies of these approaches have not been systematically assessed. We found that most standard approaches for fitting and removing mean task-evoked activations were unable to correct these inflated correlations. In contrast, methods that flexibly fit mean task-evoked response shapes effectively corrected the inflated correlations without reducing effects of interest. Results with empirical fMRI data confirmed the model's predictions, revealing activation-induced task-state FC inflation for both Pearson correlation and psychophysiological interaction (PPI) approaches. These results demonstrate that removal of mean task-evoked activations using an approach that flexibly models task-evoked response shape is an important preprocessing step for valid estimation of task-state FC.

Resting-state network connectivity has been associated with a variety of cognitive abilities, yet it remains unclear how these connectivity properties might contribute to the neurocognitive computations underlying these abilities. We developed a new approach—information transfer mapping—to test the hypothesis that resting-state functional network topology describes the computational mappings between brain regions that carry cognitive task information. Here, we report that the transfer of diverse, task-rule information in distributed brain regions can be predicted based on estimated activity flow through resting-state network connections. Further, we find that these task-rule information transfers are coordinated by global hub regions within cognitive control networks. Activity flow over resting-state connections thus provides a large-scale network mechanism for cognitive task information transfer and global information coordination in the human brain, demonstrating the cognitive relevance of resting-state network topology.

Resting-state network connectivity has been associated with a variety of cognitive abilities, yet it remains unclear how these connectivity properties might contribute to the neurocognitive computations underlying these abilities. We designed a new approachinformation transfer mapping -to test the hypothesis that resting-state functional network topology describes the computational mappings between brain regions that carry cognitive task information. Confirming this, we found that diverse task-rule information could be predicted in held-out brain regions based on estimated activity flow through resting-state network connections. Further, we found that these task-rule information transfers were consistently coordinated by global hub regions within cognitive control networks. Activity flow over resting-state connections thus provides a large-scale network mechanism for cognitive task information transfer and global information coordination in the human brain, demonstrating the cognitive relevance of resting-state network topology.peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/101782 doi: bioRxiv preprint first posted online 2 The human brain is thought to be a distributed information processing device, with its routes of information transfer as a core feature determining its computational architecture. Many studies have utilized correlations among resting-state functional MRI (fMRI) time series to study functional connectivity (FC) in the human brain 1 (see ref. 1 for review). It remains unclear, however, if these resting-state FC routes are related to the brain's routes of information transfer. Evidence that group and individual differences in resting-state FC correlate with cognitive differences 2-4 suggests a systematic relationship between resting-state FC and cognitive information processing. However, without linking FC to information transfer, it remains unclear whether or how restingstate FC might mechanistically contribute to neurocognitive computations. Additionally, while a number of studies have shown that task information representations are distributed throughout the brain [5][6][7][8] , such studies have yet to reveal how these distributed representations are coordinated, and how information in any one brain region is used by other brain regions to produce cognitive computations 9 . Other studies investigating interdependence of brain regions during tasks (rather than during rest) have typically emphasized statistical dependencies between regional time series [10][11][12] , rather than considering the mechanistic transfer of task-relevant information content (reflected in task activation patterns 13 ) between those regions. Thus, it remains unclear whether or how the network topology described by either resting-state or task-evoked FC is relevant to the neurocognitive computations underlying task performance.Here, we provide evidence for a network mechanism unde...