Poster #17 - Age-related changes in the brain’s memory network: Specialization of the hippocampus revealed by graph theoretical analysis

• In Event: 1-005 - Poster Session 01
  In Poster Session: PS 01 Section - Biological Processes: Neuroscience and Genetics

Integrative Statement

Resting-state fMRI (rsfMRI) has provided important insight into the organization of the brain. Specifically, research utilizing rsfMRI has shown that brain regions tend to oscillate in synchrony and organize into intrinsic networks even during rest (Buckner & Vincent, 2007; Fox et al., 2005). One such network, the hippocampal memory network, has been identified in both children and adults and shown to be related to episodic memory ability (Riggins et al., 2016; Vincent et al., 2016). Despite the fact that this network has been identified in children, it remains unclear how its organization may differ across development, particularly during periods of large behavioral gains in memory ability (e.g., early childhood, Riggins, 2014). Graph theory analysis is a technique that may help address this gap, as it provides methods for understanding the changing organization of networks across development. Specifically, it can assess the integrity of the network in relation to other networks (e.g., frontoparietal network) and elucidate the role of specific brain regions within the network (Bullmore & Sporns, 2009). The goal of the present study was to utilize graph theory analysis to investigate the role of the hippocampus, a region critical for memory, within the hippocampal memory network in young children. Specifically, we investigated the integration of the hippocampus within the hippocampal memory network and segregation from other networks. 137 4- to 8-year-old healthy children took part in the current study. Children completed a T1-weighted structural MRI scan and a variation of an rsfMRI scan during which they watched Inscapes (Vanderwal et al., 2016). Functional and structural MRI data were preprocessed using the Data Processing Assistant for Resting-state fMRI Advanced Edition (DPARSF-A, v3.1) toolbox (Yan & Zang, 2010). Regions within and outside of the memory network were defined using coordinates from previous research (Chen et al., 2016; Fair et al., 2009; Spaniol et al., 2009). Graph theory analysis was carried out in R using the program igraph (Csardi & Nepusz, 2006). Two metrics were calculated to assess the role of both the right and left hippocampus in the memory network. The within-module degree z score was calculated to assess within network connections of the hippocampus and the participation coefficient was used to assess between network connections of the hippocampus to other networks. Associations between each metric and age were calculated to determine whether these metrics differed across the age range studied. Preliminary results indicated a significant bivariate correlation between the within-module degree z score associated with the right hippocampus and age (r = 0.23, p = .008). Specifically, the within-module degree z score increased with age, suggesting an increase in connections between the hippocampus and regions inside the memory network. No significant associations were observed for the participation coefficient. These results are consistent with the hypothesis that the hippocampus becomes increasingly integrated within the memory network between the ages of 4 and 8 years. This increased integration likely contributes to the specialization of the hippocampus for memory processing and, ultimately, contributes to improvements in memory typically observed in early childhood.

Authors

• Morgan Botdorf, University of Maryland, College Park
  Presenting Author

• Fengji Geng, Zhejiang University
  Non-Presenting Author

• Tracy Riggins, University of Maryland
  Non-Presenting Author