The efficiency of dynamic brain state transitions in working memory improves in youth.
Poster No:
1221
Submission Type:
Abstract Submission
Authors:
Xiaoyu Xu1,2, Zaixu Cui2
Institutions:
1State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing, China, 2Chinese Institute for Brain Research, Beijing, China
First Author:
Xiaoyu Xu
State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University|Chinese Institute for Brain Research
Beijing, China|Beijing, China
State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University|Chinese Institute for Brain Research
Beijing, China|Beijing, China
Co-Author:
Introduction:
Dynamic brain state transitions are required for working memory, such as moving from a low cognitive load to a working memory state (Braun et al., 2015). However, it remains unclear how the development of structural network topology supports the process of such dynamic brain state transitions. Network control theory (NCT) provided a powerful framework for studying how structural network topology informs and constrains functional dynamics (Gu et al., 2015). NCT can be used to quantify the energetic costs, namely control energy, required to facilitate the transition between different states. Our previous work reports control energy for frontoparietal activation decreased with development (Cui et al., 2020). In the current study, we depict how the development of structural networks facilitates the dynamic transitions in working memory execution in youth under the framework of NCT.
Methods:
The present study leveraged magnetic resonance imaging (MRI) data from the Lifespan Human Connectome Project Development (HCP-D) study (Somerville et al., 2018). The final sample included 590 subjects aged 8-22 with complete anatomical and diffusion MRI (dMRI) data. We preprocessed dMRI data and reconstruct the structural networks using the pipeline integrated within qsiprep (Cieslak et al., 2021). The functional activation map for the n-back task was extracted from the Human Connectome Project (Glasser et al., 2016). To depict the transition from 0-back to 2-back, we defined the initial brain states as all zero, and the target state as the contrast map between 2-back and 0-back conditions. Based on NCT, we calculated the control energy required by all brain regions during the transition (Fig.1A). To validate the facilitation of structural network topology on brain state transition, we compared the control energy of the actual networks to the null networks with the same degree and strength distribution.
Next, we used general additive models to assess the developmental effects of control energy at the levels of the whole brain, systems, and nodes. Multiple comparisons were accounted for using the False Discovery Rate (Q<0.05). All the developmental effects were compared to the effects observed in the null networks. Furthermore, we described the developmental patterns among different systems leveraging the first derivative. We also correlated the age effect sizes of nodes to the sensorimotor-association axis (S-A axis) (Sydnor et al., 2021) using the Spearman method and spin test (Alexander-Bloch et al., 2018).
Next, we used general additive models to assess the developmental effects of control energy at the levels of the whole brain, systems, and nodes. Multiple comparisons were accounted for using the False Discovery Rate (Q<0.05). All the developmental effects were compared to the effects observed in the null networks. Furthermore, we described the developmental patterns among different systems leveraging the first derivative. We also correlated the age effect sizes of nodes to the sensorimotor-association axis (S-A axis) (Sydnor et al., 2021) using the Spearman method and spin test (Alexander-Bloch et al., 2018).
Results:
The frontoparietal system exhibits the highest energy consumption in facilitating the transition of brain states from 0-back to 2-back (Fig.1B, C). The energetic cost of real structural networks is significantly lower than that of null networks (t=-175.14, P<2.2e-16, Fig.1D). The average energetic cost required for the whole brain declines with development (P=3.53e-09, r=-0.25, Fig2.A), and the age effect size of real networks is significantly stronger than that of null networks. As to the system level, both the average energetic costs of the frontoparietal and somatomotor systems exhibit strong developmental effects. However, the decreasing of energetic cost is consistent in the frontoparietal system from childhood to adulthood, while the cost of the somatomotor system hits a plateau in the middle of adolescence (Fig.2B, C, D). Furthermore, the age effect size (Z stats) of nodes tends to converge towards the S-A axis (r=0.28, Pspin=0.004, Fig.2E, F). This tendency reflects a hierarchical pattern of development of structural network topology.
Conclusions:
Our results reveal how the development of structural network topology supports the dynamic brain state transitions in working memory. The theoretical energy consumption declined during development, which illustrates that the efficiency of dynamic brain state transitions in working memory improves from childhood to adulthood.
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making 2
Lifespan Development:
Early life, Adolescence, Aging 1
Lifespan Development Other
Keywords:
Cognition
Development
MRI
White Matter
WHITE MATTER IMAGING - DTI, HARDI, DSI, ETC
Other - executive function
1|2Indicates the priority used for review
·Figure 1. Schematic of methods.
·Figure 2. Energetic costs in n-back task change with age.
Provide references using author date format
Braun, U. (2015). Dynamic reconfiguration of frontal brain networks during executive cognition in humans. Proc Natl Acad Sci U S A, 112(37), 11678-11683.
Cieslak, M. (2021). QSIPrep: an integrative platform for preprocessing and reconstructing diffusion MRI data. Nat Methods, 18(7), 775-778.
Cui, Z. (2020). Optimization of energy state transition trajectory supports the development of executive function during youth. Elife, 9.
Glasser, M. F. (2016). The Human Connectome Project's neuroimaging approach. Nat Neurosci, 19(9), 1175-1187.
Gu, S. (2015). Controllability of structural brain networks. Nat Commun, 6, 8414.
Somerville, L. H. (2018). The Lifespan Human Connectome Project in Development: A large-scale study of brain connectivity development in 5-21 year olds. Neuroimage, 183, 456-468.
Sydnor, V. J. (2021). Neurodevelopment of the association cortices: Patterns, mechanisms, and implications for psychopathology. Neuron, 109(18), 2820-2846.