Non-linear development of EEG coherence in adolescents and young adults shown by the analysis of neurophysiological trajectories and their covariance.
- Authors
- Chorlian, David B; Kamarajan, Chella; Meyers, Jacquelyn L; Pandey, Ashwini K; Zhang, Jian; Kinreich, Sivan; Porjesz, Bernice
- Year
- 2024
- Journal
- bioRxiv : the preprint server for biology
- PMID
- 38559025
- DOI
- 10.1101/2024.03.13.584867
- PMCID
- PMC10980032
To contribute to the understanding of changes in the factors governing the development of neural connectivity, the developmental structure of EEG coherence in adolescents and young adults was analyzed using the means, variances, and covariances of high alpha frequency band coherence measures from a set of 27 coherence pairs obtained from a sample of 1426 participants from the COGA study with 5006 observations over ages 12 through 31. Means and covariances were calculated at 96 age centers by a LOESS method. In the current study, trajectories of covariance matrices considered as individual units were determined by tensorial analysis: calculation of Riemannian geodesic (non-Euclidean) distances between matrices and application of both linear and non-linear dimension reduction techniques to these distances. Results were evaluated by bootstrap methods. Mean coherence trajectories for males and females were very similar, showing a steady upward trend from ages 12 to 20 which diminishes gradually from 20 to 25 and reaches stability from 25 to 31. In contrast, the individual covariance trajectories of males and female differed, with the male covariance levels becoming greater than that of females during the developmental process. Tensorial determination of the distances from the initial covariance matrix of subsequent covariance matrices to age 20 had the same trajectory as the mean coherence values. Tensorial determination of the trajectories of the covariance matrices of males and females based on their all pairs geodesic distances revealed a non-linear pattern in the multi-dimensional space of each of the trajectories: A steady increase in one dimension is accompanied by deviations from it peaking at age 20 which have both transient and lasting effects. There is a precise temporal parallelism of this pattern of covariance in males and females, while there is a consistent distance between the male and female covariance structures throughout the developmental process. Between region differences (anterior-posterior) within each sex are greater than between sex differences within regions. Examining development using multiple methods provides unique insight into the developmental process.
Upper panel: Frontal-central electrodes and sagittal bipolar derivations. Blue lines connect the electrodes forming the bipolar channels. Lower panel: Blue lines connect the electrodes forming the bipolar channels. Red lines connect midpoints of coherence pairs for all pairs calculations. Central-parietal is similar.
LLM interpretation
This figure consists of two schematic diagrams illustrating electrode placements and connectivity for EEG analysis. The upper panel shows sagittal frontal-central bipolar derivations, where blue lines connect specific electrode pairs (e.g., F7-T7, F3-C3). The lower panel displays coherence pairs, maintaining the blue bipolar channels while adding red lines that connect the midpoints of these pairs to form various coherence combinations.
Geodesic linearity analysis of entire system.
LLM interpretation
This figure consists of two line graphs comparing geodesic linearity analysis for males (left) and females (right), with the x-axis representing age in years and the y-axis representing geodesic distance. Both plots track four metrics: primary linear progression (solid blue), secondary linear progression (solid green), distance from primary progression (dashed blue), and distance from secondary progression (dashed green). In both groups, the primary linear progression increases steadily with age, while the distance from primary progression peaks around age 20 before declining.
Geodesic linearity analysis of Anterior and Posterior subsystems.
LLM interpretation
This figure consists of four line graphs showing geodesic linearity analysis for Male and Female subjects in the Anterior and Posterior subsystems. The x-axis represents "Age in years" (15–30) and the y-axis represents "Geodesic distance." Each plot compares primary and secondary linear progressions (solid lines) against their respective distances from progression (dashed lines), with the primary linear progression generally showing an upward trend as age increases.
Derivations and coherence pairs for lateral intra- and interhemispheric coherence pairs. Interhemispheric pairs are not used in this analysis.
LLM interpretation
This is a schematic diagram illustrating laterally oriented coherence pairs across the brain's hemispheres. The figure displays intrahemispheric pairs (connected by solid lines) and interhemispheric pairs (connected by dashed lines) between specific electrode sites (e.g., F7-F3, P7-P3). Red curved lines group these pairs into lateral clusters, while blue circles and red diamonds mark the specific derivation points.
High Alpha band trajectories of the 27 bipolar coherence pairs used in this study. Note that there is little difference between male and female trajectories.
LLM interpretation
This figure consists of six line graphs showing coherence trajectories across age (12 to 31 years) for male and female groups, categorized by brain region: anterior, posterior, and intra. Each plot displays multiple colored lines representing 27 bipolar coherence pairs, with the y-axis ranging from 0 to 1. The trajectories generally show a slight increase in coherence during adolescence followed by a plateau, with similar patterns observed between males and females.
Z-scores of anterior and posterior high alpha band trajectories of the 27 bipolar coherence pairs used in this study. Coherence pairs were divided into 2 similar groups for each region: an inter-hemispheric group of 4 coherence pairs, and an intra-hemispheric and midline group of 6 coherence pairs. See Section 6.1 for complete lists. Male trajectories are represented by solid lines and female trajectories by dashed lines. The blue and cyan lines in the upper right panel are F7-C7 – F3-C3 and F7-C7 – FZ-CZ. Intrahemispheric pairs F7-F3 – T7-C3 and F7-F3 – P7-P4 also show anomalous trajectories not illustrated here.
LLM interpretation
This figure consists of four line plots showing Z-scores of high alpha band trajectories across age (12 to 31 years) for anterior-inter, anterior-intra, posterior-inter, and posterior-intra coherence pairs. Solid lines represent male trajectories and dashed lines represent female trajectories, with various colors denoting different coherence pairs. While most trajectories show a general increase in Z-score with age, the "Anterior-Intra" panel displays a distinct divergence where several trajectories (notably blue and cyan) peak around age 20 and then decline.
Upper Panel: Trajectories of the quartiles of the covariance values. Lower Panel: Trajectories of the quartiles of the correlation values.
LLM interpretation
This figure consists of two line plots showing the trajectories of covariance values (upper panel) and correlation coefficients (lower panel) from age 12 to 31. Both panels compare males (solid lines) and females (dashed lines) across the 1st quartile, median, and 3rd quartile. In both visualizations, values generally trend upward over time, with the 3rd quartile (red) consistently remaining the highest and the 1st quartile (blue) the lowest for both sexes.
The z-score of the geodesic distance between the initial (age 12) covariance matrix of the entire set of 27 coherence pairs and each successive covariance matrix was superimposed on the Z-scores of anterior and posterior high alpha band trajectories of the 27 bipolar coherence pairs used in this study. Coherence pairs were divided into 2 similar groups for each region: an inter-hemispheric group of 4 coherence pairs, and an intra-hemispheric and midline group of 6 coherence pairs. See section 6.1 for complete lists. Male trajectories are represented by solid lines and female trajectories by dashed lines. The blue and cyan lines in the upper right panel are F7-C7 – F3-C3 and F7-C7 – FZ-CZ. Intrahemispheric pairs F7-F3 – T7-C3 and F7-F3 – P7-P4 also show anomalous trajectories not illustrated here. Geodesic distance trajectories are represented by black lines.
LLM interpretation
This figure consists of four line plots showing z-score trajectories from age 12 to 31 for anterior-inter, anterior-intra, posterior-inter, and posterior-intra coherence pairs. The x-axis represents age in years, and the y-axis represents the z-score, with solid lines indicating males, dashed lines indicating females, and black lines representing geodesic distance trajectories. Most trajectories show an upward trend that plateaus after age 20, though the "Anterior-Intra" panel displays several divergent trajectories (blue and cyan) that decline after age 20.
MDS trajectories for the complete covariance matrices.
LLM interpretation
This line graph displays Multidimensional Scaling (MDS) distance values across three axes for males (solid lines) and females (dashed lines) from ages 12 to 30. The x-axis represents "Age in years" and the y-axis represents "MDS distance." Axis 1 (blue) shows a steady increase over time, while Axis 2 (green) and Axis 3 (red) exhibit oscillatory patterns, with male and female trajectories closely following one another.
MDS trajectories for region/sex comparisons. Upper left: Anterior Male-Female comparison. Male lines are solid, Female dashed. Upper right: Male Anteror-Posterior comparison. Anterior lines are solid, Posterior dashed. Lower left: Posterior Male-Female comparison. Male lines are solid, Female dashed. Lower right: Female Anteror-Posterior comparison. Anterior lines are solid, Posterior dashed.
LLM interpretation
This figure consists of four line plots showing Multidimensional Scaling (MDS) distance trajectories across age (12 to 31 years) for different sex and region comparisons. The x-axis represents "Age in years" and the y-axis represents "MDS distance," with colored lines (red, green, blue) representing different data dimensions. Solid lines denote males or anterior regions, while dashed lines denote females or posterior regions, as specified by the panel titles.
Trajectories of Riemannian distances between males and females by region and between regions within sexes.
LLM interpretation
This line graph displays geodesic distances between sexes and brain regions across ages 12 to 31 years. The y-axis represents "Geodesic distances" and the x-axis represents "Age in years." The "Male-Female distance" (solid blue line) shows the highest overall values and a general upward trend, while other trajectories, including sex-specific anterior-posterior distances and regional sex differences, remain relatively stable between values of 1.0 and 2.3.
No entities extracted from this document yet.
No uploaded files.
| Citation | PMID | DOI | Status |
|---|---|---|---|
| AgrawalA, BrislinSJ, BucholzKK, DickD, HartRP, JohnsonEC, MeyersJ, SalvatoreJ, SlesingerP, COGA Collaborators; AlmasyL, ForoudT, GoateA, HesselbrockV, KramerJ, KupermanS, MerikangasAK, NurnbergerJI, TischfieldJ, EdenbergHJ, PorjeszB. The Collaborative Study on the Genetics of Alcoholism: Overview. Genes Brain Behav. 2023 Oct;22(5):e12864. doi: 10.1111/gbb.12864. Epub 2023 Sep 22.37736010 PMC10550790 | — | — | — |
| ArsignyV., FillardP., PennecX., AyacheN. (2006) Log-Euclidean Metrics for Fast and Simple Calculus on Diffusion Tensors. Magnetic Resonance in Medicine 56:411–42116788917 10.1002/mrm.20965 | — | — | — |
| BegleiterH., ReichT., HesselbrockV.M., PorjeszB., LiT.K., SchuckitM.A., , (1995) The Collaborative Study on the Genetics of Alcoholism. Alcohol Health Res. World 19:228–236.31798102 PMC6875768 | — | — | — |
| ChorlianDB, MeyersJL, ManzN, ZhangJ, KamarajanC, PandeyA, WangJC, PlaweckiM, EdenbergH, GoateA, TischfieldJ, PorjeszB. Genetic influences vary by age and sex: Trajectories of the association of cholinergic system variants and theta band event related oscillations. bioRxiv [Preprint]. 2023 Feb 28:2023.02.27.530318. doi: 10.1101/2023.02.27.530318. | — | — | — |
| ChorlianDB, RangaswamyM, ManzN, KamarajanC, PandeyAK, EdenbergH, KupermanS, PorjeszB. (2015) Gender modulates the development of Theta Event Related Oscillations in Adolescents and Young Adults. Behav Brain Res. 292:342–5226102560 10.1016/j.bbr.2015.06.020PMC4705839 | — | — | — |
| ChorlianDB, RangaswamyM, ManzN, KamarajanC, PandeyAK, WangJC, WetherillL, EdenbergH, PorjeszB. (2017) Genetic correlates of the development of Theta Event Related Oscillations in Adolescents and Young Adults. Int J Psychophysiol. 115:24–3927847216 10.1016/j.ijpsycho.2016.11.007PMC5456461 | — | — | — |
| ChorlianDB, RangaswamyM, PorjeszB. EEG coherence: topography and frequency structure. Exp Brain Res. 2009 Sep;198(1):59–83.19626316 10.1007/s00221-009-1936-9 | — | — | — |
| CousminerDL, BerryDJ, TimpsonNJ, AngW, ThieringE, ByrneEM, TaalHR, HuikariV, BradfieldJP, KerkhofM, Groen-BlokhuisMM, Kreiner-M Kreiner-MøllerE, MarinelliM, HolstC, LeinonenJT, PerryJR, SurakkaI, PietiläinenO, KettunenJ, AnttilaV, KaakinenM, SovioU, PoutaA, DasS, LagouV, PowerC, ProkopenkoI, EvansDM, KempJP, St PourcainB, RingS, PalotieA, KajantieE, OsmondC, LehtimäkiT, ViikariJS, KähönenM, WarringtonNM, LyeSJ, PalmerLJ, TieslerCM, FlexederC, MontgomeryGW, MedlandSE, HofmanA, HakonarsonH, GuxensM, BartelsM, SalomaaV; ReproGen Consortium, MurabitoJM, KaprioJ, SørensenTI, BallesterF, BisgaardH, BoomsmaDI, KoppelmanGH, GrantSF, JaddoeVW, MartinNG, HeinrichJ, PennellCE, RaitakariOT, ErikssonJG, SmithGD, HyppönenE, JärvelinMR, McCarthyMI, RipattiS, WidénE; Early Growth Genetics (EGG) Consortium. (2013) Genome-wide association and longitudinal analyses reveal genetic loci linking pubertal height growth, pubertal timing and childhood adiposity. Hum Mol Genet. 22(13):2735–47.23449627 10.1093/hmg/ddt104PMC3674797 | — | — | — |
| DrydenI. L.,, AlexeyKoloydenko, DiweiZhou, (2009) Non-euclidean statistics for covariance matrices, with applications to diffusion tensor imaging The Annals of Applied Statistics, 3:3, 1102–1123 2009 | — | — | — |
| FletcherP. T., JoshiS. (2007) Riemannian Geometry for the Statistical Analysis of Diffusion Tensor Data. Signal Processing 87:2, 250–262 2007 | — | — | — |
| FletcherP. T., LuC., PizerS. M. and JoshiS. (2004) Principal geodesic analysis for the study of nonlinear statistics of shape. IEEE Transactions on Medical Imaging, 23:8, 995–1005, 200415338733 10.1109/TMI.2004.831793 | — | — | — |
| Meyers JLD. B., ChorlianT. B., BigdeliE. C., JohnsonF., AlievA., AgrawalL., AlmasyA., AnokhinH. J., EdenbergT., ForoudA., GoateC., KamarajanS., KinreichJ., NurnbergerA. K., PandeyG., PandeyM. H., PlaweckiJ. E., SalvatoreJ., ZhangA., FanousB., Porjesz (2021) The association of polygenic risk for schizophrenia, bipolar disorder, and depression with neural connectivity in adolescents and young adults: examining developmental and sex differences Translational Psychiatry 2021 11(54)10.1038/s41398-020-01185-7PMC780946233446638 | — | — | — |
| MeyersJL, ChorlianD., JohnsonE., PandeyA., KamarajanC., SalvatoreJ., AlievF., Subbie-Saenz de ViteriS., ZhangJ., ChaoM., KapoorM, HesselbrockV., KramerJ., KupermanS., NurnbergerJ., TischfieldJ., GoateA., ForoudT., DickD., EdenbergH., AgrawalA., PorjeszB. (2019) Association of Polygenic Liability for Alcohol Dependence and EEG Connectivity in Adolescence and Young Adulthood Brain Sci. 2019 Oct 17;9(10).10.3390/brainsci9100280PMC682673531627376 | — | — | — |
| MeyersJL, ZhangJ, ChorlianDB, PandeyAK, KamarajanC, WangJC, WetherillL, LaiD, ChaoM, ChanG, KinreichS, KapoorM, BertelsenS, McClintickJ, BauerL, HesselbrockV, KupermanS, KramerJ, SalvatoreJE, DickDM, AgrawalA, ForoudT, EdenbergHJ, GoateA, PorjeszB. (2020) A genome-wide association study of interhemispheric theta EEG coherence: implications for neural connectivity and alcohol use behavior Mol Psychiatry. 2020 May 20.10.1038/s41380-020-0777-6PMC850386032433515 | — | — | — |
| MitteroeckerP., BooksteinF. (2009) The ontogenetic trajectory of the phenotypic covariance matrix, with examples from craniofacial shape in rats and humans. Evolution, 63, 727–737. 200919087182 10.1111/j.1558-5646.2008.00587.x | — | — | — |
| MitteroeckerP., BooksteinF. (2014) Comparing Covariance Matrices by Relative Eigenanalysis, with Applications to Organismal Biology. Evol Biol 41:336–350 2014 | — | — | — |
| PennecX., FillardP. AyacheN., A Riemannian Framework for Tensor Computing. RR-5255, INRIA. 2004, pp.34. inria-00070743 | — | — | — |
| SteadJD, NealC, MengF, WangY, EvansS, VazquezDM, WatsonSJ, AkilH. (2006) Transcriptional profiling of the developing rat brain reveals that the most dramatic regional differentiation in gene expression occurs postpartum. J Neurosci 26(1):345–53.16399705 10.1523/JNEUROSCI.2755-05.2006PMC6674315 | — | — | — |
No papers in this knowledge base cite this source.
No citations found for this source.