Modularity of Brain Networks for Egocentric and Allocentric Memory-guided Reaching

2079 

Abstract Submission 

Lina Musa¹, Amirhossein Ghaderi¹, Ying Chen², J. Douglas Crawford¹

¹York University, Toronto, Ontario, ²Queen's University, Kingston, Ontario

Lina Musa  
York University
Toronto, Ontario

Amirhossein Ghaderi  
York University
Toronto, Ontario

Ying Chen  
Queen's University
Kingston, Ontario

J. Douglas Crawford  
York University
Toronto, Ontario

The brain can encode targets for reaching in egocentric and/or allocentric reference frames (Byrne and Crawford 2010). The differences in the cortical activation of these two representations has been described (Chen et al., 2014; Neggers et al., 2006). For example, Chen et al. (2014) identified egocentric directional selectivity in dorsal brain areas (the parieto-frontal cortex) versus landmark-centered directional selectivity in ventral brain areas (inferior temporal gyrus and inferior occipital gyrus) during a delayed reach task. However, differences in the functional organization of brain networks have not been studied.

Here, we performed a secondary analysis of the event-related fMRI task from Chen et al. (2014), to distinguish human brain networks involved in egocentric versus allocentric spatial representation of reach targets. Based on their previous univariate analysis we expected that the functional brain networks will differ, with increased hubness in ventral brain regions in the allocentric task. The paradigm consisted of three tasks with identical stimulus display but different instructions: egocentric reach (remember absolute target location), allocentric reach (remember target location relative to a visual landmark), and a nonspatial control, color report (report color of target). We performed a graph theoretical analysis on time series data recorded during the memory delay period, contrasting egocentric and allocentric data versus baseline and control. Network hubs, clustering coefficient, and efficiency of the networks were found. The community organization of the network into modules was determined using the Newman's spectral community detection approach (Newman, 2006) and the consensus partitioning of participant data. Dynamical measures of network connectivity, the synchrony and complexity, of network modules were quantified using the energy and Shannon entropy, respectively.

Both the egocentric (Figure-1) and allocentric (Figure-2) brain networks showed increased functional segregation & integration, relative to control. In both tasks, there were no inferotemporal modules, rather the data were largely segregated into occipito-dorsal-parietal and & temporo-frontal networks modules, with similar organization in egocentric vs. allocentric trials. Contrary to expectations, the allocentric network demonstrated significantly stronger modularity in the occipito-dorsal-parietal module relative to the egocentric network, although it did demonstrate increased connectivity between modules as compared to the egocentric brain network. In addition, the allocentric network showed an increase of intramodular hubs (brain regions that were important for within module information transfer) and intermodular hubs (brain regions that were important for sharing information between modules) in the occipito-dorsal-parietal module. Lastly, for the allocentric network, there was increase in desynchronization and complexity in the occipito-dorsal-parietal module, relative to the egocentric network, indicating an increase in difficulty of information processing.

Our results demonstrate that rather than increased allocentric encoding of visual reach targets in the ventral stream, there is increased specialization in the interaction between early visual brain areas and dorsal parietal brain areas. This potentially demonstrates an importance of the dorsal parietal cortex in allocentric spatial encoding of visuomotor targets, through the integration visual information about task stimuli with object spatial information in the parietal cortex. Patients with allocentric spatial neglect similarly demonstrate an importance in the spatial encoding of objects in the parietal cortex. Specialization, however, is accompanied with integration and increased interaction with the temporo-frontal network, which likely facilitates the processing of instructions of using a landmark.

fMRI Connectivity and Network Modeling ²

Motor Planning and Execution

Visuo-Motor Functions ¹

FUNCTIONAL MRI

Vision

Other - Graph Theory