State space modeling of time-varying contemporaneous and lagged relations in connectivity maps

Most connectivity mapping techniques for neuroimaging data assume stationarity (i.e., network parameters are constant across time), but this assumption does not always hold true. The authors provide a description of a new approach for simultaneously detecting time-varying (or dynamic) contemporaneous and lagged relations in brain connectivity maps. Specifically, they use a novel raw data likelihood estimation technique (involving a second-order extended Kalman filter/smoother embedded in a nonlinear optimizer) to determine the variances of the random walks associated with state space model parameters and their autoregressive components. The authors illustrate their approach with simulated and blood oxygen level-dependent functional magnetic resonance imaging data from 30 daily cigarette smokers performing a verbal working memory task, focusing on seven regions of interest (ROIs). Twelve participants had dynamic directed functional connectivity maps: Eleven had one or more time-varying contemporaneous ROI state loadings, and one had a time-varying autoregressive parameter. Compared to smokers without dynamic maps, smokers with dynamic maps performed the task with greater accuracy. Thus, accurate detection of dynamic brain processes is meaningfully related to behavior in a clinical sample.

[1]  J. Fiez,et al.  Neural correlates of self-focused and other-focused strategies for coping with cigarette cue exposure. , 2013, Psychology of addictive behaviors : journal of the Society of Psychologists in Addictive Behaviors.

[2]  Karl J. Friston,et al.  Analysing connectivity with Granger causality and dynamic causal modelling , 2013, Current Opinion in Neurobiology.

[3]  Olaf Sporns,et al.  Synchronization dynamics and evidence for a repertoire of network states in resting EEG , 2012, Front. Comput. Neurosci..

[4]  Vesa Kiviniemi,et al.  A Sliding Time-Window ICA Reveals Spatial Variability of the Default Mode Network in Time , 2011, Brain Connect..

[5]  Lutz Kilian,et al.  NEW INTRODUCTION TO MULTIPLE TIME SERIES ANALYSIS, by Helmut Lütkepohl, Springer, 2005 , 2006, Econometric Theory.

[6]  Li Hu,et al.  A time-varying source connectivity approach to reveal human somatosensory information processing , 2012, NeuroImage.

[7]  Maciej Niedzwiecki,et al.  Identification of Time-Varying Processes , 2000 .

[8]  David A. Leopold,et al.  Dynamic functional connectivity: Promise, issues, and interpretations , 2013, NeuroImage.

[9]  D. Simon.  Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches , 2006 .

[10]  D. Borsboom.  Measuring the mind: Conceptual issues in contemporary psychometrics , 2005 .

[11]  M. Arnold,et al.  Instantaneous multivariate EEG coherence analysis by means of adaptive high-dimensional autoregressive models , 2001, Journal of Neuroscience Methods.

[12]  Kathleen M. Gates,et al.  Automatic search for fMRI connectivity mapping: An alternative to Granger causality testing using formal equivalences among SEM path modeling, VAR, and unified SEM , 2010, NeuroImage.

[13]  Daniel A. Handwerker,et al.  Periodic changes in fMRI connectivity , 2012, NeuroImage.

[14]  Enzo Tagliazucchi,et al.  Dynamic BOLD functional connectivity in humans and its electrophysiological correlates , 2012, Front. Hum. Neurosci..

[15]  Thia Kirubarajan,et al.  Estimation with Applications to Tracking and Navigation: Theory, Algorithms and Software , 2001 .

[16]  Laura Astolfi,et al.  A new Kalman filter approach for the estimation of high-dimensional time-variant multivariate AR models and its application in analysis of laser-evoked brain potentials , 2010, NeuroImage.

[17]  R W Cox,et al.  AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. , 1996, Computers and biomedical research, an international journal.

[18]  Dimitri P. Bertsekas,et al.  Nonlinear Programming , 1997 .

[19]  SY-MIIN CHOW,et al.  Nonlinear Regime-Switching State-Space (RSSS) Models , 2013, Psychometrika.

[20]  Helmut Ltkepohl,et al.  New Introduction to Multiple Time Series Analysis , 2007 .

[21]  Karin Schwab,et al.  A Time-Variant Processing Approach for the Analysis of Alpha and Gamma MEG Oscillations During Flicker Stimulus Generated Entrainment , 2011, IEEE Transactions on Biomedical Engineering.

[22]  Kathleen M. Gates,et al.  Extended unified SEM approach for modeling event-related fMRI data , 2011, NeuroImage.

[23]  Kewei Chen,et al.  The Accounting Review , 1972 .

[24]  Siem Jan Koopman,et al.  Time Series Analysis by State Space Methods , 2001 .

[25]  Eswar Damaraju,et al.  Tracking whole-brain connectivity dynamics in the resting state. , 2014, Cerebral cortex.

[26]  Peter C M Molenaar,et al.  Statistical Modeling of the Individual: Rationale and Application of Multivariate Stationary Time Series Analysis , 2005, Multivariate behavioral research.

[27]  Peter C. M. Molenaar,et al.  A posteriori model validation for the temporal order of directed functional connectivity maps , 2015, Front. Neurosci..

[28]  P. Gill,et al.  User's Guide for SOL/NPSOL: A Fortran Package for Nonlinear Programming. , 1983 .

[29]  Kathleen M. Gates,et al.  Group search algorithm recovers effective connectivity maps for individuals in homogeneous and heterogeneous samples , 2012, NeuroImage.

[30]  Nelson J. Trujillo-Barreto,et al.  A switching multi-scale dynamical network model of EEG/MEG , 2013, NeuroImage.

[31]  Xiao Liu,et al.  EEG correlates of time-varying BOLD functional connectivity , 2013, NeuroImage.

[32]  Qing X. Yang,et al.  Networks involved in olfaction and their dynamics using independent component analysis and unified structural equation modeling , 2014, Human brain mapping.

[33]  Peter C. M. Molenaar,et al.  A dynamic factor model for the analysis of multivariate time series , 1985 .

[34]  Hillary D. Schwarb,et al.  Short‐time windows of correlation between large‐scale functional brain networks predict vigilance intraindividually and interindividually , 2013, Human brain mapping.

[35]  Timothy E. J. Behrens,et al.  Human connectomics , 2012, Current Opinion in Neurobiology.

[36]  D. Paré,et al.  Contrasting Activity Profile of Two Distributed Cortical Networks as a Function of Attentional Demands , 2009, The Journal of Neuroscience.

[37]  Vince D. Calhoun,et al.  Dynamic modeling of neuronal responses in fMRI using cubature Kalman filtering , 2011, NeuroImage.

[38]  Y. Bar-Shalom.  Tracking and data association , 1988 .

[39]  Moriah E. Thomason,et al.  Vector autoregression, structural equation modeling, and their synthesis in neuroimaging data analysis , 2011, Comput. Biol. Medicine.

[40]  Wei Zhu,et al.  Unified structural equation modeling approach for the analysis of multisubject, multivariate functional MRI data , 2007, Human brain mapping.

[41]  Carsten Thomsen,et al.  A comparative study between a simplified Kalman filter and Sliding Window Averaging for single trial dynamical estimation of event-related potentials , 2010, Comput. Methods Programs Biomed..

[42]  Otto W. Witte,et al.  Modelling and analysis of time-variant directed interrelations between brain regions based on BOLD-signals , 2009, NeuroImage.

[43]  Thomas T. Liu,et al.  Caffeine increases the temporal variability of resting-state BOLD connectivity in the motor cortex , 2012, NeuroImage.

[44]  Kent A. Kiehl,et al.  A method for evaluating dynamic functional network connectivity and task-modulation: application to schizophrenia , 2010, Magnetic Resonance Materials in Physics, Biology and Medicine.

[45]  Catie Chang,et al.  Time–frequency dynamics of resting-state brain connectivity measured with fMRI , 2010, NeuroImage.

[46]  Mark S. Cohen,et al.  Effects of acute smoking on brain activity vary with abstinence in smokers performing the N-Back Task: A preliminary study , 2006, Psychiatry Research: Neuroimaging.

[47]  Julie A Fiez,et al.  Quitting-unmotivated and quitting-motivated cigarette smokers exhibit different patterns of cue-elicited brain activation when anticipating an opportunity to smoke. , 2012, Journal of abnormal psychology.

[48]  M. Corbetta,et al.  Temporal dynamics of spontaneous MEG activity in brain networks , 2010, Proceedings of the National Academy of Sciences.

[49]  Mark W. Woolrich,et al.  Network modelling methods for FMRI , 2011, NeuroImage.

[50]  E. Stein,et al.  Chronic smoking, but not acute nicotine administration, modulates neural correlates of working memory , 2010, Psychopharmacology.

[51]  Graham C. Goodwin,et al.  Adaptive filtering prediction and control , 1984 .

[52]  Giorgio E. Primiceri.  Time Varying Structural Vector Autoregressions and Monetary Policy , 2002 .

[53]  Stephen M. Smith,et al.  The future of FMRI connectivity , 2012, NeuroImage.

[54]  J. Polzehl,et al.  Structural Adaptive Smoothing Procedures , 2008 .

[55]  Peter C M Molenaar,et al.  Greater BOLD activity but more efficient connectivity is associated with better cognitive performance within a sample of nicotine‐deprived smokers , 2014, Addiction biology.

[56]  David T. Jones,et al.  Non-Stationarity in the “Resting Brain’s” Modular Architecture , 2012, PloS one.

[57]  Phil Wood.  Confirmatory Factor Analysis for Applied Research , 2008 .