Research reportGranger causal connectivity dissociates navigation networks that subserve allocentric and egocentric path integration
Introduction
Successful navigation in well-known and unknown environments requires simultaneous processing and integration of spatial information based on allocentric and egocentric spatial reference frames (SRFs) (Klatzky, 1998). Reference frames are a means to represent spatial information based on egocentric or allocentric coordinate systems. An allocentric representational system is centered on aspects of the environment and represents the location of entities in space with respect to allothetic information like cardinal directions. In contrast, an egocentric representational system is centered on aspects of the navigator’s physical structure and thus varies with changes in orientation of the navigator. Importantly, successful navigation requires integration of spatial information from both egocentric and allocentric representations to allow goal-directed action in the environment (Gramann, 2013).
The computation, integration, and exchange of spatial information based different SRFs involves a network of brain structures including the medial temporal cortex, the cingulate gyrus, the frontal, parietal, and occipital cortices, as well as the retrosplenial complex (RSC) (Hartley et al., 2003, Maguire et al., 1998, Whitlock et al., 2008). Imaging studies investigating the neural structures underlying egocentric and allocentric spatial navigation have revealed that the parietal cortex subserves the computation of egocentric SRFs by integrating self-motion cues from the kinesthetic, vestibular, and visual systems (Zaehle et al., 2007, Committeri et al., 2004, Cohen and Andersen, 2002). In contrast, the use of an allocentric SRF mainly engages medial temporal brain structures (Doeller et al., 2010, Ekstrom et al., 2003, Howard et al., 2014, Jacobs et al., 2013, Maguire et al., 1998, Wolbers and Büchel, 2005). Moreover, the RSC has been found to play important roles in computing and maintaining allocentric spatial representations and in transforming spatial information between egocentric and allocentric reference frames (Byrne et al., 2007, Dhindsa et al., 2014, Vann et al., 2009, Zhang et al., 2012).
Many of these brain areas are simultaneously active during navigation tasks, and coupling of functionally specialized brain regions appears to be necessary for successful navigation (Ekstrom et al., 2014). Recent EEG studies have reported high coherence of the alpha and theta frequency bands in a large-scale cortical network recruited during spatial navigation (Li et al., 2009, Ramos-Loyo and Sanchez-Loyo, 2011). Connectivity across various brain areas with modulations in the theta and alpha frequency ranges may support the synchronization of large-scale cortical interactions (Palva and Palva, 2011, Sauseng et al., 2005) and is one of the essential neuronal mechanisms for higher cognitive functions (Siegel et al., 2012). However, investigations describing the flow of information within these cortical networks with high temporal resolution are scarce, and the architecture of the spatial navigation network is not well understood.
To further our understanding of connectivity in the navigation network, we used high-density EEG and Granger causality analysis to investigate which brain regions are causally connected while participants updated their position and orientation during navigation. Previous studies using path integration paradigms showed that the individual preference to use either an egocentric or an allocentric reference frame is stable for individuals (Gramann et al., 2005), is based on higher cognitive functions (Gramann et al., 2009), depends on core areas of the navigation network (Gramann et al., 2006, Gramann et al., 2010, Seubert et al., 2008), and can be reliably observed in different populations (Gramann et al., 2012, Goeke et al., 2013, Goeke et al., 2015). Previous studies also demonstrated navigation-related modulations of distinct frequency bands that were dependent on the reference frame proclivity of participants (Chiu et al., 2012, Gramann et al., 2010, Lin et al., 2015, Plank et al., 2010). To further investigate the information flow in the human navigation network and to understand how information flow differs between egocentric and allocentric navigators, we analyzed granger causal information flow in EEG data recorded during a virtual path integration task.
Section snippets
Results
For allocentric and egocentric participants, the behavioral performance including homing angle and homing position were reported. The analysis of direct information transfer between clusters of ICs revealed event related causality (ERC) in the time-frequency distribution between several cortical regions. Widespread brain regions were involved in path integration, revealing directed ERC between the anterior cingulate cortex (ACC), the RSC, and the lateral prefrontal, motor, parietal, and
Discussion
In this study, we found ERC flow in the delta (1–3.5 Hz), theta (4–7 Hz), alpha (8–13 Hz), and beta (14–30 Hz) frequency bands in the human navigation network during virtual path integration. The dominant frequency characteristics of this network were in line with previous EEG studies demonstrating theta power increases in the frontal cortex to co-vary with alpha power changes in the motor, parietal, and occipital cortices as well as the RSC (Lin et al., 2015, Chiu et al., 2012, Plank et al.,
Limitation and conclusion
Our work discovered a large-scale navigation network subserving spatial orientation on the basis of egocentric and allocentric SRFs. A number of caveats need to be noted regarding the current research. The first limitation is the data analysis. Since the signal-to-noise ratio is poor in EEG data, the relative processing methods are needed to extract the useful information from EEG. The analysing methods including pre-processing, noise removal, source separation and location were also required
Homing task
We used a VR path integration task with passive transportation during environments with clear geometric structure and rich visual flow information (for a detailed description of the task please see Lin et al., 2015). Participants always started from the same position (marked by star in Fig. 6A) in the VR scenario and were passively guided along different trajectories (as shown in Fig. 6). All trajectories were composed of varying numbers of straight segments of the same length before and after
Acknowledgments
This work was supported in part by the Australian Research Council (ARC) under discovery grant DP180100670. This research was also sponsored in part by the Army Research Laboratory and was accomplished under Cooperative Agreement Number W911NF-10-2-0022 and W911NF-10-D-0002/TO 0023.
References (79)
- et al.
Estimation of the cortical functional connectivity with the multimodal integration of high-resolution EEG and fMRI data by directed transfer function
Neuroimage
(2005) - et al.
Dorsolateral prefrontal contributions to human working memory
Cortex
(2013) Prefrontal and parietal contributions to spatial working memory
Neuroscience
(2006)- et al.
Persistent activity in the prefrontal cortex during working memory
Trends Cogn. Sci.
(2003) - et al.
EEGLAB: An open source toolbox for analysis of single-trial EEG dynamics including independent component analysis
J. Neurosci. Methods
(2004) - et al.
The neural basis of ego- and allocentric reference frames in spatial navigation: evidence from spatio-temporal coupled current density reconstruction
Brain Res.
(2006) - et al.
The well-worn route and the path less traveled: distinct neural bases of route following and wayfinding in humans
Neuron
(2003) - et al.
The hippocampus and entorhinal cortex encode the path and euclidean distances to goals during navigation
Curr. Biol.
(2014) - et al.
Directional disorientation following left retrosplenial hemorrhage: a case report with fMRI studies
Cortex
(2007) - et al.
Determination of information flow direction among brain structures by a modified directed transfer function (dDTF) method
J. Neurosci. Methods
(2003)