Longitudinal modeling of age-dependent latent traits with generalized additive latent and mixed models

We present generalized additive latent and mixed models (GALAMMs) for analysis of clustered data with latent and observed variables depending smoothly on observed variables. A profile likelihood algorithm is proposed, and we derive asymptotic standard errors of both smooth and parametric terms. The work was motivated by applications in cognitive neuroscience, and we show how GALAMMs can successfully model the complex lifespan trajectory of latent episodic memory, along with a discrepant trajectory of working memory, as well as the effect of latent socioeconomic status on hippocampal development. Simulation experiments suggest that model estimates are accurate even with moderate sample sizes.

[1]  Eric J. Pedersen,et al.  Hierarchical generalized additive models in ecology: an introduction with mgcv , 2019, PeerJ.

[2]  Christina Gloeckner,et al.  Modern Applied Statistics With S , 2003 .

[3]  Soroosh Afyouni,et al.  Confound modelling in UK Biobank brain imaging , 2020, NeuroImage.

[4]  William Meredith,et al.  Latent curve analysis , 1990 .

[5]  J. Mcardle,et al.  Comparative longitudinal structural analyses of the growth and decline of multiple intellectual abilities over the life span. , 2002, Developmental psychology.

[6]  F. Vaida,et al.  Conditional Akaike information for mixed-effects models , 2005 .

[7]  K. Jöreskog.  A general approach to confirmatory maximum likelihood factor analysis , 1969 .

[8]  Simon N. Wood,et al.  Generalized Additive Models for Gigadata: Modeling the U.K. Black Smoke Network Daily Data , 2017 .

[9]  X. Lin,et al.  Inference in generalized additive mixed modelsby using smoothing splines , 1999 .

[10]  J. Ware,et al.  Random-effects models for longitudinal data. , 1982, Biometrics.

[11]  G. Molenberghs,et al.  Mixed models approaches for joint modeling of different types of responses , 2016, Journal of biopharmaceutical statistics.

[12]  Klaus P. Ebmeier,et al.  Education and Income Show Heterogeneous Relationships to Lifespan Brain and Cognitive Differences Across European and US Cohorts , 2020, bioRxiv.

[13]  Minjeong Jeon,et al.  Estimating Complex Measurement and Growth Models Using the R Package PLmixed , 2019, Multivariate behavioral research.

[14]  H. Swaminathan,et al.  Detecting Differential Item Functioning Using Logistic Regression Procedures , 1990 .

[15]  P. Ghisletta,et al.  A dynamic investigation of cognitive dedifferentiation with control for retest: evidence from the Swiss Interdisciplinary Longitudinal Study on the Oldest Old. , 2005, Psychology and aging.

[16]  T. Salthouse.  When does age-related cognitive decline begin? , 2009, Neurobiology of Aging.

[17]  John A. Nelder,et al.  A Simplex Method for Function Minimization , 1965, Comput. J..

[18]  Holger Brandt,et al.  A Nonlinear Structural Equation Mixture Modeling Approach for Nonnormally Distributed Latent Predictor Variables , 2014 .

[19]  Edgar C. Merkle,et al.  merDeriv: Derivative Computations for Linear Mixed Effects Models with Application to Robust Standard Errors , 2016, 1612.04911.

[20]  G. Wahba,et al.  A Correspondence Between Bayesian Estimation on Stochastic Processes and Smoothing by Splines , 1970 .

[21]  F. Ostrosky-solis,et al.  Digit Span: Effect of education and culture , 2006 .

[22]  Jarrod Had.  MCMC Methods for Multi-Response Generalized Linear Mixed Models: The MCMCglmm R Package , 2010 .

[23]  Eric Kan,et al.  Neural correlates of socioeconomic status in the developing human brain. , 2012, Developmental science.

[24]  L. Fahrmeir,et al.  A Bayesian Semiparametric Latent Variable Model for Mixed Responses , 2007 .

[25]  Ron Dumont,et al.  California Verbal Learning Test, Second Edition , 2014 .

[26]  A. Tröster,et al.  Test-retest stability of the California verbal learning test in older persons. , 1997, Neuropsychology.

[27]  Minjeong Jeon,et al.  Profile-Likelihood Approach for Estimating Generalized Linear Mixed Models With Factor Structures , 2012 .

[28]  Edward H. Ip,et al.  A Bayesian Modeling Approach for Generalized Semiparametric Structural Equation Models , 2013, Psychometrika.

[29]  K. Walhovd,et al.  Within-session verbal learning slope is predictive of lifespan delayed recall, hippocampal volume, and memory training benefit, and is heritable , 2020, Scientific Reports.

[30]  Michael Moutoussis,et al.  Developmental cognitive neuroscience using latent change score models: A tutorial and applications , 2017, Developmental Cognitive Neuroscience.

[31]  U. Lindenberger,et al.  Coupled Cognitive Changes in Adulthood: A Meta-Analysis , 2019, Psychological bulletin.

[32]  B. Muthén.  BEYOND SEM: GENERAL LATENT VARIABLE MODELING , 2002 .

[33]  Kelvin K. W. Yau,et al.  Conditional Akaike information criterion for generalized linear mixed models , 2012, Comput. Stat. Data Anal..

[34]  Education and Income Show Heterogeneous Relationships to Lifespan Brain and Cognitive Differences Across European and US Cohorts , 2021, Cerebral cortex.

[35]  William R. Parke.  Pseudo Maximum Likelihood Estimation: The Asymptotic Distribution , 1986 .

[36]  Eugene Demidenko,et al.  Mixed Models: Theory and Applications with R , 2013 .

[37]  J J McArdle,et al.  Latent growth curves within developmental structural equation models. , 1987, Child development.

[38]  Roger Sauter,et al.  In All Likelihood , 2002, Technometrics.

[39]  É. Moulines,et al.  Convergence of a stochastic approximation version of the EM algorithm , 1999 .

[40]  Anders M. Dale,et al.  When does brain aging accelerate? Dangers of quadratic fits in cross-sectional studies , 2010, NeuroImage.

[41]  Jorge Nocedal,et al.  A Limited Memory Algorithm for Bound Constrained Optimization , 1995, SIAM J. Sci. Comput..

[42]  ohn,et al.  Additive models with predictors subject to measurement error , 2004 .

[43]  G. Molenberghs.  Applied Longitudinal Analysis , 2005 .

[44]  Benoit Liquet,et al.  Estimation of extended mixed models using latent classes and latent processes: the R package lcmm , 2015, 1503.00890.

[45]  Michael C Neale,et al.  People are variables too: multilevel structural equations modeling. , 2005, Psychological methods.

[46]  Bruce Fischl,et al.  Within-subject template estimation for unbiased longitudinal image analysis , 2012, NeuroImage.

[47]  E. B. Wilson.  Probable Inference, the Law of Succession, and Statistical Inference , 1927 .

[48]  L. Nyberg,et al.  Stability, growth, and decline in adult life span development of declarative memory: cross-sectional and longitudinal data from a population-based study. , 2005, Psychology and aging.

[49]  B. Muthén.  A general structural equation model with dichotomous, ordered categorical, and continuous latent variable indicators , 1984 .

[50]  Louise Ryan,et al.  Bivariate Latent Variable Models for Clustered Discrete and Continuous Outcomes , 1992 .

[51]  Donald R. Lynam,et al.  The quandary of covarying: A brief review and empirical examination of covariate use in structural neuroimaging studies on psychological variables , 2020, NeuroImage.

[52]  D. Bates,et al.  Nonlinear mixed effects models for repeated measures data. , 1990, Biometrics.

[53]  Andreas Brezger,et al.  Generalized structured additive regression based on Bayesian P-splines , 2006, Comput. Stat. Data Anal..

[54]  P. Baltes.  Longitudinal and cross-sectional sequences in the study of age and generation effects. , 1968, Human development.

[55]  Tom A. B. Snijders,et al.  Multilevel Analysis , 2011, International Encyclopedia of Statistical Science.

[56]  H. Akaike.  A new look at the statistical model identification , 1974 .

[57]  P. A. V. B. Swamy,et al.  Statistical Inference in Random Coefficient Regression Models , 1971 .

[58]  L. Whalley,et al.  Childhood socioeconomic status and adult brain size: Childhood socioeconomic status influences adult hippocampal size , 2012, Annals of neurology.

[59]  P. Hall,et al.  Theory for penalised spline regression , 2005 .

[60]  H. Holling,et al.  Retest effects in cognitive ability tests: A meta-analysis , 2018 .

[61]  D. Bates,et al.  Fitting Linear Mixed-Effects Models Using lme4 , 2014, 1406.5823.

[62]  M. Wand,et al.  Simple Incorporation of Interactions into Additive Models , 2001, Biometrics.

[63]  Robin Thompson,et al.  [That BLUP is a Good Thing: The Estimation of Random Effects]: Comment , 1991 .

[64]  Holger Brandt,et al.  A general non-linear multilevel structural equation mixture model , 2014, Front. Psychol..

[65]  Anders M. Fjell,et al.  A recipe for accurate estimation of lifespan brain trajectories, distinguishing longitudinal and cohort effects , 2021, NeuroImage.

[66]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[67]  I. Priede,et al.  Improving the precision of the daily egg production method using generalized additive models , 1997 .

[68]  S. Wood,et al.  Smoothing Parameter and Model Selection for General Smooth Models , 2015, 1511.03864.

[69]  Daniel J. Bauer.  A Semiparametric Approach to Modeling Nonlinear Relations Among Latent Variables , 2005 .

[70]  N. Breslow,et al.  Approximate inference in generalized linear mixed models , 1993 .

[71]  R. Tibshirani,et al.  Generalized Additive Models , 1986 .

[72]  Mary J. Lindstrom,et al.  Penalized spline models for longitudinal data , 2008 .

[73]  M. R. Novick.  The axioms and principal results of classical test theory , 1965 .

[74]  M. Kenward,et al.  The Analysis of Designed Experiments and Longitudinal Data by Using Smoothing Splines , 1999 .

[75]  P. Mair.  Path Analysis and Structural Equation Models , 2018 .

[76]  D. A. Kenny,et al.  Estimating the nonlinear and interactive effects of latent variables. , 1984 .

[77]  C. Jack,et al.  Preclinical Alzheimer's disease: Definition, natural history, and diagnostic criteria , 2016, Alzheimer's & Dementia.

[78]  Douglas Nychka,et al.  Bayesian Confidence Intervals for Smoothing Splines , 1988 .

[79]  Alan Y. Chiang,et al.  Generalized Additive Models: An Introduction With R , 2007, Technometrics.

[80]  H. Wickham.  Simple, Consistent Wrappers for Common String Operations , 2015 .

[81]  Charles C. Driver,et al.  Hierarchical Bayesian continuous time dynamic modeling. , 2018, Psychological methods.

[82]  S. Rabe-Hesketh,et al.  Generalized multilevel structural equation modeling , 2004 .

[83]  Quincy M. Samus,et al.  Dementia prevention, intervention, and care , 2017, The Lancet.

[84]  Jeffrey B. Arnold.  Extra Themes, Scales and Geoms for 'ggplot2' , 2016 .

[85]  Nadja Klein,et al.  Bayesian structured additive distributional regression for multivariate responses , 2015 .

[86]  Geert Verbeke,et al.  Joint models for high-dimensional longitudinal data , 2008 .

[87]  Thomas Lumley,et al.  Exposure and measurement contributions to estimates of acute air pollution effects , 2005, Journal of Exposure Analysis and Environmental Epidemiology.

[88]  Hadley Wickham,et al.  ggplot2 - Elegant Graphics for Data Analysis (2nd Edition) , 2017 .

[89]  Paul-Christian Bürkner,et al.  Advanced Bayesian Multilevel Modeling with the R Package brms , 2017, R J..

[90]  B. Muthén,et al.  A Bayesian approach to nonlinear latent variable models using the Gibbs sampler and the metropolis-hastings algorithm , 1998 .

[91]  B. Silverman,et al.  Some Aspects of the Spline Smoothing Approach to Non‐Parametric Regression Curve Fitting , 1985 .

[92]  Marie Davidian,et al.  Non-linear mixed-effects models , 2008 .

[93]  E. Tucker-Drob.  Cognitive Aging and Dementia: A Life Span Perspective. , 2019, Annual review of developmental psychology.

[94]  Cécile Proust-Lima,et al.  Analysis of multivariate mixed longitudinal data: a flexible latent process approach. , 2012, The British journal of mathematical and statistical psychology.

[95]  Mark Tiede,et al.  Investigating dialectal differences using articulography , 2016, ICPhS.

[96]  Geert Verbeke,et al.  Pairwise Fitting of Mixed Models for the Joint Modeling of Multivariate Longitudinal Profiles , 2006, Biometrics.

[97]  Computation and application of generalized linear mixed model derivatives using lme4 , 2020, Psychometrika.

[98]  Kenneth A. Bollen,et al.  Latent curve models: A structural equation perspective , 2005 .

[99]  B. Muthén.  Latent variable structural equation modeling with categorical data , 1983 .

[100]  L. E. Jones,et al.  Analysis of multiplicative combination rules when the causal variables are measured with error. , 1983 .

[101]  Fabian Scheipl,et al.  Straightforward intermediate rank tensor product smoothing in mixed models , 2012, Statistics and Computing.

[102]  S Y Lee,et al.  Statistical analysis of nonlinear structural equation models with continuous and polytomous data. , 2000, The British journal of mathematical and statistical psychology.

[103]  F. Bowman,et al.  Spatiotemporal Models for Region of Interest Analyses of Functional Neuroimaging Data , 2007 .

[104]  R. Newcombe.  Two-sided confidence intervals for the single proportion: comparison of seven methods. , 1998, Statistics in medicine.

[105]  U. Lindenberger,et al.  Only time will tell: cross-sectional studies offer no solution to the age-brain-cognition triangle: comment on Salthouse (2011). , 2011, Psychological bulletin.

[106]  Andrew W. Roddam,et al.  Measurement Error in Nonlinear Models: a Modern Perspective , 2008 .

[107]  Xiangnan Feng,et al.  Latent variable models with nonparametric interaction effects of latent variables , 2014, Statistics in medicine.

[108]  A. Daugherty,et al.  Socioeconomic status and hippocampal volume in children and young adults. , 2018, Developmental science.

[109]  D. Dunson,et al.  Bayesian Semiparametric Structural Equation Models with Latent Variables , 2010 .

[110]  Tx Station.  Stata Statistical Software: Release 7. , 2001 .

[111]  S. Wood,et al.  Coverage Properties of Confidence Intervals for Generalized Additive Model Components , 2012 .

[112]  K. Warner Schaie,et al.  “When does age-related cognitive decline begin?” Salthouse again reifies the “cross-sectional fallacy” , 2009, Neurobiology of Aging.

[113]  John C. Nash,et al.  Unifying Optimization Algorithms to Aid Software System Users: optimx for R , 2011 .

[114]  L. Nyberg,et al.  Memory aging and brain maintenance , 2012, Trends in Cognitive Sciences.

[115]  Matthew N. O. Sadiku,et al.  General Intelligence , 2021, A Primer on Multiple Intelligences.

[116]  S. Wood.  Low‐Rank Scale‐Invariant Tensor Product Smooths for Generalized Additive Mixed Models , 2006, Biometrics.

[117]  K. Jöreskog.  A General Method for Estimating a Linear Structural Equation System. , 1970 .

[118]  D. Delis,et al.  The California Verbal Learning Test--second edition: test-retest reliability, practice effects, and reliable change indices for the standard and alternate forms. , 2006, Archives of clinical neuropsychology : the official journal of the National Academy of Neuropsychologists.

[119]  J. Horn,et al.  A practical and theoretical guide to measurement invariance in aging research. , 1992, Experimental aging research.

[120]  F. Dominici,et al.  On the use of generalized additive models in time-series studies of air pollution and health. , 2002, American journal of epidemiology.

[121]  Xinyuan Song,et al.  Semiparametric Latent Variable Models With Bayesian P-Splines , 2010 .

[122]  Stanley R. Johnson,et al.  Varying Coefficient Models , 1984 .

[123]  S. Wood.  Thin plate regression splines , 2003 .

[124]  Nicholas J. Rockwood,et al.  Maximum Likelihood Estimation of Multilevel Structural Equation Models with Random Slopes for Latent Covariates , 2020, Psychometrika.

[125]  John C. Nash,et al.  On Best Practice Optimization Methods in R , 2014 .

[126]  Deborah G. Mayo,et al.  Statistical Inference as Severe Testing , 2018 .

[127]  S. Greven,et al.  On the behaviour of marginal and conditional AIC in linear mixed models , 2010 .

[128]  Stine K. Krogsrud,et al.  Neurodevelopmental origins of lifespan changes in brain and cognition , 2016, Proceedings of the National Academy of Sciences.

[129]  S. Rabe-Hesketh,et al.  Latent Variable Modelling: A Survey * , 2007 .

[130]  K Y Liang,et al.  An overview of methods for the analysis of longitudinal data. , 1992, Statistics in medicine.

[131]  J. Hintze,et al.  Violin plots : A box plot-density trace synergism , 1998 .

[132]  E. Ferrer,et al.  Studying developmental processes in accelerated cohort-sequential designs with discrete- and continuous-time latent change score models. , 2019, Psychological methods.

[133]  Marc Lavielle,et al.  Parameter Estimation in Nonlinear Mixed Effect Models Using saemix, an R Implementation of the SAEM Algorithm , 2017 .

[134]  Jeffrey R. Harring,et al.  Nonlinear Mixed-Effects Modeling Programs in R , 2018 .

[135]  L. Nyberg,et al.  Challenging the notion of an early-onset of cognitive decline , 2009, Neurobiology of Aging.

[136]  Simon N. Wood,et al.  Inference and computation with generalized additive models and their extensions , 2020, TEST.

[137]  Raymond J. Carroll,et al.  Measurement error in nonlinear models: a modern perspective , 2006 .

[138]  K. Blennow,et al.  Neuroinflammation and Tau Interact with Amyloid in Predicting Sleep Problems in Aging Independently of Atrophy , 2018, Cerebral cortex.

[139]  L. Ryan,et al.  Latent Variable Models for Mixed Discrete and Continuous Outcomes , 1997 .

[140]  S. Rabe-Hesketh,et al.  Maximum likelihood estimation of limited and discrete dependent variable models with nested random effects , 2005 .

[141]  Eric R. Ziegel,et al.  Generalized Linear Models , 2002, Technometrics.

[142]  Pamela K. Smith,et al.  Models of visuospatial and verbal memory across the adult life span. , 2002, Psychology and aging.

[143]  B. Ripley,et al.  Semiparametric Regression: Preface , 2003 .

[144]  J. Oud,et al.  Continuous time state space modeling of panel data by means of sem , 2000 .

[145]  S. Wood,et al.  Generalized additive models for large data sets , 2015 .

[146]  S. Greven,et al.  A unifying approach to the estimation of the conditional Akaike information in generalized linear mixed models , 2014 .

[147]  Paul-Christian Bürkner,et al.  brms: An R Package for Bayesian Multilevel Models Using Stan , 2017 .

[148]  Klaus P. Ebmeier,et al.  Educational attainment does not influence brain aging , 2021, Proceedings of the National Academy of Sciences.

[149]  R. Schall.  Estimation in generalized linear models with random effects , 1991 .

[150]  Mert R. Sabuncu,et al.  Spatiotemporal Linear Mixed Effects Modeling for the Mass-univariate Analysis of Longitudinal Neuroimage Data ⁎ for the Alzheimer's Disease Neuroimaging Initiative 1 , 2022 .

[151]  S. Pollak,et al.  Association between Income and the Hippocampus , 2011, PloS one.

[152]  A. Benton,et al.  Revised administration and scoring of the digit span test. , 1957, Journal of consulting psychology.

[153]  T. Salthouse.  Neuroanatomical substrates of age-related cognitive decline. , 2011, Psychological bulletin.

[154]  Gail Gong,et al.  Pseudo Maximum Likelihood Estimation: Theory and Applications , 1981 .

[155]  H. Westervelt,et al.  Practice effects, test-retest stability, and dual baseline assessments with the California Verbal Learning Test in an HIV sample. , 2001, Archives of clinical neuropsychology : the official journal of the National Academy of Neuropsychologists.

[156]  S. Wood.  Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models , 2011 .

[157]  E. Vonesh,et al.  A note on the use of Laplace's approximation for nonlinear mixed-effects models , 1996 .

[158]  S. Wood.  Stable and Efficient Multiple Smoothing Parameter Estimation for Generalized Additive Models , 2004 .

[159]  S G West,et al.  Putting the individual back into individual growth curves. , 2000, Psychological methods.

[160]  David Ruppert,et al.  Theory & Methods: Spatially‐adaptive Penalties for Spline Fitting , 2000 .

[161]  R. Gueorguieva,et al.  A multivariate generalized linear mixed model for joint modelling of clustered outcomes in the exponential family , 2001 .

[162]  R. Bagozzi,et al.  On the nature and direction of relationships between constructs and measures. , 2000, Psychological methods.

[163]  Robert J. Mislevy,et al.  Estimation of Latent Group Effects , 1985 .

[164]  Wm. R. Wright.  General Intelligence, Objectively Determined and Measured. , 1905 .

[165]  Daniel J. Bauer,et al.  The disaggregation of within-person and between-person effects in longitudinal models of change. , 2011, Annual review of psychology.

[166]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[167]  Patrick J Curran,et al.  Have Multilevel Models Been Structural Equation Models All Along? , 2003, Multivariate behavioral research.

[168]  Anders M. Dale,et al.  Cortical Surface-Based Analysis I. Segmentation and Surface Reconstruction , 1999, NeuroImage.

[169]  S. Lang,et al.  Bayesian P-Splines , 2004 .

[170]  C. Wild,et al.  Vector Generalized Additive Models , 1996 .