Effects of Development and Cognitive Training on Brain Signal Variability
Poster No:
1107
Submission Type:
Abstract Submission
Authors:
Fan LIU1, Feiyan CHEN1
Institutions:
1Bio‑X Laboratory, School of Physics, Zhejiang University, HangZhou, China
First Author:
Co-Author:
Introduction:
Previous works on both fMRI(Grady and Garrett 2014) and EEG(Mcintosh, Kovacevic, and Itier 2008) suggested that the variability in brain signal is necessary for individuals to adapt to the changing environment. Moreover, the brain signal variability shows an "inverted U-shaped"(Grady and Garrett 2014) dynamic trajectory, where young adults with better cognitive abilities have the greatest brain signal variability(Mcintosh et al. 2010). Therefore, the metric could be a potential biomarker of aging and changing of cognitive abilities.
However, previous studies were based on task-fMRI(Boylan et al. 2021; Grady and Garrett 2014) or defined the fixation before tasks as the resting data(Garrett et al. 2011), while the fixation is more likely to be influenced by the preceding and following tasks. In addition, previous studies have focused on aging adults or younger adults (20-30 years old)(Garrett et al. 2011) to explore the changes in aging. Finally, there was a large age span of subjects in the same study, which may confound some of the underlying changes.
Therefore, our study used the resting-state fMRI and select first, fourth, and sixth graders in primary school to delineate these developmental changes in brain signal variability through school years. Additionally, abacus-based mental calculation training (AMC) has been used as a cognitive training to study the effects of training intervention on variability.
However, previous studies were based on task-fMRI(Boylan et al. 2021; Grady and Garrett 2014) or defined the fixation before tasks as the resting data(Garrett et al. 2011), while the fixation is more likely to be influenced by the preceding and following tasks. In addition, previous studies have focused on aging adults or younger adults (20-30 years old)(Garrett et al. 2011) to explore the changes in aging. Finally, there was a large age span of subjects in the same study, which may confound some of the underlying changes.
Therefore, our study used the resting-state fMRI and select first, fourth, and sixth graders in primary school to delineate these developmental changes in brain signal variability through school years. Additionally, abacus-based mental calculation training (AMC) has been used as a cognitive training to study the effects of training intervention on variability.
Methods:
We recruited a group of children of whom resting-state fMRI (rs-fMRI) were collected in Grade 1, Grade 4 (three years' AMC training) and Grade 6 (five years' AMC training), as well as the matched control group.
Preprocessing steps included discarding the first five images, slice-timing, and head-motion correction. Then the functional images were aligned to the corresponding T1-weighted images, and were normalized to the MNI space with a resampling voxel size of 3 × 3 × 3 mm3. The spatial smoothing was skipped to reduce the impact of signals from other voxels(Garrett et al. 2010). The scans with excessive head motion (3 mm and 3°) were excluded. Some nuisance variables were removed in multiple linear regression analysis, including 24 Friston body-motion parameters and average white matter, cerebrospinal fluid. And a band-pass (0.01–0.1 Hz) filter was applied to reduce the effects of physiological noise. Then we extracted the time series for ROIs defined by the BN-246 template(Fan et al. 2016). MSE (multiscale sample entropy) were computed for the variability of signals.
Preprocessing steps included discarding the first five images, slice-timing, and head-motion correction. Then the functional images were aligned to the corresponding T1-weighted images, and were normalized to the MNI space with a resampling voxel size of 3 × 3 × 3 mm3. The spatial smoothing was skipped to reduce the impact of signals from other voxels(Garrett et al. 2010). The scans with excessive head motion (3 mm and 3°) were excluded. Some nuisance variables were removed in multiple linear regression analysis, including 24 Friston body-motion parameters and average white matter, cerebrospinal fluid. And a band-pass (0.01–0.1 Hz) filter was applied to reduce the effects of physiological noise. Then we extracted the time series for ROIs defined by the BN-246 template(Fan et al. 2016). MSE (multiscale sample entropy) were computed for the variability of signals.
Results:
We found that brain signal variability presented an "U-shaped" dynamic trajectory during school years, which is different from the previous studies. The trend of variability was firstly decrease and then an increase occurred after the fourth grade, which could be accelerated by AMC training (see Fig 1).
In the ROI-level analysis, we observed that after three years of AMC training, the AMC group already showed a significant decrease in brain signal variability in the orbital gyrus, inferior temporal gyrus, basal ganglia, and thalamus compared to that of first year. The variability then shifted to an elevated trend with continued training. In addition, more brain regions in the control group remained in a state of reduced variability at grade 6 compared to AMC group. In grade 4, the AMC group showed lower variability compared to the control group, while in grade 6, the AMC group had higher brain signal variability. Thus, these results revealed that training accelerated the dynamical change of brain signal variability. The AMC group achieves the transition from lower to higher brain signal variability earlier than children in the control group.
In the ROI-level analysis, we observed that after three years of AMC training, the AMC group already showed a significant decrease in brain signal variability in the orbital gyrus, inferior temporal gyrus, basal ganglia, and thalamus compared to that of first year. The variability then shifted to an elevated trend with continued training. In addition, more brain regions in the control group remained in a state of reduced variability at grade 6 compared to AMC group. In grade 4, the AMC group showed lower variability compared to the control group, while in grade 6, the AMC group had higher brain signal variability. Thus, these results revealed that training accelerated the dynamical change of brain signal variability. The AMC group achieves the transition from lower to higher brain signal variability earlier than children in the control group.
·Fig 1. The MSE results of the brain signal variability.
Conclusions:
Brain signal variability can be used as a biomarker of children's development and training. It exhibited an "U-shaped" dynamic trajectory among the school-aged children. Notably, the intervention of AMC training accelerates this dynamic process, making the metric reach a minimum more quickly and then return to increasing.
Brain Stimulation:
Non-Invasive Stimulation Methods Other
Learning and Memory:
Skill Learning 1
Modeling and Analysis Methods:
Task-Independent and Resting-State Analysis 2
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
Cognition
Development
FUNCTIONAL MRI
NORMAL HUMAN
Other - Brain signal variability
1|2Indicates the priority used for review
Provide references using author date format
Fan, Lingzhong.(2016), “The Human Brainnetome Atlas: A New Brain Atlas Based on Connectional Architecture.” Cerebral Cortex 26(8): 3508–26.
Garrett. (2010), “Blood Oxygen Level-Dependent Signal Variability Is More than Just Noise.” The Journal of Neuroscience 30(14): 4914.
Garrett. (2011). “The Importance of Being Variable.” The Journal of Neuroscience 31(12): 4496-503.
Grady. (2014). “Understanding Variability in the BOLD Signal and Why It Matters for Aging.” Brain Imaging and Behavior pages274–283.
Mcintosh, A R. (2010), “The Development of a Noisy Brain.” Archives Italiennes de Biologie 148(3): 323–37.
Mcintosh, A R. (2008), “Increased Brain Signal Variability Accompanies Lower Behavioral Variability in Development.” PLoS Comput Biol 4(7): 1000106.