Poster No:
1805
Submission Type:
Abstract Submission
Authors:
Diana Perez Rivera1, Gretchen Wulfekuhle2, Joanna Hernandez1, Evan Gordon3, Caterina Gratton2,1
Institutions:
1Northwestern University, Evanston, IL, 2Florida State University, Tallahassee, FL, 3Washington University, St. Louis, MO
First Author:
Co-Author(s):
Caterina Gratton
Florida State University|Northwestern University
Tallahassee, FL|Evanston, IL
Introduction:
The brain is divided into multiple large-scale distributed systems or "networks". Network interactions within and across systems give rise to cognition and behavior. Past research has suggested that these interactions change systematically with age, with networks becoming relatively desegregated from one another (Chan et al., 2014). However, individuals vary widely in the topology of brain networks (Gordon et al., 2017), particularly of higher-order association systems (Seitzman et al., 2019). Yet, most existing research uses a priori network partitions based on a group average (typically from young adult data) to extract measures of functional networks. This approach, while convenient, can introduce further noise by combining signals from different networks when the individual networks deviate from the group average. Obtaining reliable representations of individual-specific network architecture, however, requires large amounts of data per subject; typically more data than is collected in most studies. Here, we compare measures of within- and between-network FC, and a derivative measure of network segregation in highly sampled younger and older adults, using both group-based and individualized networks.
Methods:
We leveraged two precision fMRI datasets of younger (n = 16; ages 18-30) and older (n=8; ages 65-75) adults with >90 min. of high-quality rs-FC data per subject. Networks were defined in two ways: 1) based on a priori group-based partition (Gordon et al., 2016), and 2) based on each individual's rs-FC patterns (Gordon et al., 2017). The large amount of data per individual allowed us to produce highly-reliable representations of individual-specific networks. For each network partition, we extracted values for each individual of within-network FC and of between-network FC, and we calculated a segregation index. To determine how these metrics are influenced by individual differences in network topology, we used two-way ANOVAs with age-group and network partition as factors.
Results:
Our results indicate that within-network FC shows a main effect of age-group (p(FDR)<0.013) and of network partition (p(FDR)<0.013). Younger adults showed greater within-network FC compared to older adults and within-network FC was higher when calculated based on individualized networks. Similarly, between-network FC showed a significant main effect of network partition (p(FDR)<0.013), where between-network FC was significantly higher when based on a group-based networks compared to individualized networks. Network segregation showed both a main effect of age-group (p(FDR)<0.013) and of network partition (p(FDR)<0.013). Older adults showed lower segregation index compared to younger adults, replicating previous results, and this pattern was present across both network partitions, though segregation indices were significantly higher for individualized networks.
Conclusions:
Our results suggest that FC metrics in both younger and older adults are influenced by individual differences in network topology. In particular, within-network FC and its derivative, network segregation, differ significantly when calculated based on group-based networks compared to individual-specific networks. We attribute this difference to poor fit of group-average networks to individual brains that leads to non-homogenous FC measures.
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling 1
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Cortical Anatomy and Brain Mapping 2
Keywords:
ADULTS
Cortex
FUNCTIONAL MRI
Other - Networks
1|2Indicates the priority used for review
Provide references using author date format
Chan, M. Y., et al. (2014). Decreased segregation of brain systems across the healthy adult lifespan. Proceedings of the National Academy of Sciences, 111(46).
Gordon, E. M., et al. (2016). Generation and Evaluation of a Cortical Area Parcellation from Resting-State Correlations. Cerebral Cortex, 26(1), 288–303.
Gordon, E. M., et al. (2017). Precision Functional Mapping of Individual Human Brains. Neuron, 95(4), 791-807.e7.
Seitzman, B. A., Gratton, C., et al. (2019). Trait-like variants in human functional brain networks. Proceedings of the National Academy of Sciences, 116(45), 22851–22861.