A5060765720- Mental Health Research Topics
- Statistical Methods and Bayesian Inference
- Child and Adolescent Psychosocial and Emotional Development
- Statistical Methods and Inference
- Behavioral Health and Interventions
- Complex Network Analysis Techniques
- Functional Brain Connectivity Studies
- Infant Health and Development
- Advanced Statistical Modeling Techniques
- Digital Mental Health Interventions
- Health disparities and outcomes
- Sensory Analysis and Statistical Methods
- Attachment and Relationship Dynamics
- Cognitive Science and Mapping
- Early Childhood Education and Development
- Complex Systems and Decision Making
- Psychological Well-being and Life Satisfaction
- Control Systems and Identification
- Complex Systems and Time Series Analysis
- Maternal Mental Health During Pregnancy and Postpartum
- demographic modeling and climate adaptation
- Cognitive Abilities and Testing
- Bayesian Methods and Mixture Models
- Opinion Dynamics and Social Influence
- Child Abuse and Trauma
Pennsylvania State University
2016-2025
Masaryk University
2024
National Center for Advancing Translational Sciences
2021
U.S. National Science Foundation
2021
Utrecht University
2017
University of North Carolina at Chapel Hill
2007-2016
Texas Tech University
2016
University of New Mexico
2015
Health and Human Development (2HD) Research Network
2004-2013
Social Science Research Council
2011
The authors present in this study a damped oscillator model that provides direct mathematical basis for testing the notion of emotion as self-regulatory thermostat. Parameters from reflect individual differences emotional lability and ability to regulate emotion. discuss concepts such intensity, rate change, acceleration context emotion, they illustrate strengths approach comparison with spectral analysis growth curve models. utility modeling is illustrated using daily ratings 179 college...
This study determined the relative efficacy of an experimental explicit emergent literacy intervention program for preschoolers experiencing multiple risk factors. Using alternating treatment research design, children completed two 6-week waves in small groups; one wave featured program, whereas other a comparison program. Emergent assessment was conducted at pretest and end each wave. Results indicated significant widespread gains knowledge over entire 12-week program; growth significantly...
Automated facial measurement using computer vision has the potential to objectively document continuous changes in behavior. To examine emotional expression and communication, we used automated measurements quantify smile strength, eye constriction, mouth opening two 6‐month‐old infant‐mother dyads who each engaged a face‐to‐face interaction. showed high associations with anatomically based manual coding (concurrent validity); of smiling mean ratings positive emotion made by naive observers...
One of the promises experience sampling methodology (ESM) is that a statistical analysis an individual's emotions, cognitions and behaviors in everyday-life could be used to identify relevant treatment targets. A requisite for clinical implementation outcomes such person-specific time-series analyses are not wholly contingent on researcher performing them. To evaluate this, we crowdsourced one individual patient's ESM data 12 prominent research teams, asking them what symptom(s) they would...
State-space modeling techniques have been compared to structural equation (SEM) in various contexts but their unique strengths often overshadowed by similarities SEM. In this article, we provide a comprehensive discussion of these 2 approaches' and differences through analytic comparisons numerical simulations, with focus on use representing intraindividual dynamics interindividual differences. To demonstrate the respective weaknesses approaches aspects, simulated data under (a)...
Self-regulation can be conceptualized in terms of dynamic tension between highly probable reactions (prepotent responses) and use strategies that modulate those (executive processes). This study investigated the value a dynamical systems approach to early childhood self-regulation. Specifically, ordinary differential equations (ODEs) were used model interactive influences 115 36-month-olds' executive processes (strategy use) prepotent responses waiting open gift (desire for frustration about...
Self-regulation is a dynamic process wherein executive processes (EP) delay, minimize or desist prepotent responses (PR) that arise in situations threaten well-being. It generally assumed that, over the course of early childhood, children expand and more effectively deploy their repertoire EP-related strategies to regulate PR. However, longitudinal tests these assumptions are scarce part because self-regulation has been mostly studied as static construct. This study engages systems modeling...
Factor analysis is a popular statistical technique for multivariate data analysis. Developments in the structural equation modeling framework have enabled use of hybrid confirmatory/exploratory approaches which factor-loading structures can be explored relatively flexibly within confirmatory factor (CFA) framework. Recently, Muthén & Asparouhov proposed Bayesian (BSEM) approach to explore presence cross loadings CFA models. We show that issue determining patterns may formulated as variable...
Intensive longitudinal data in the behavioral sciences are often noisy, multivariate nature, and may involve multiple units undergoing regime switches by showing discontinuities interspersed with continuous dynamics. Despite increasing interest using linear nonlinear differential/difference equation models switches, there has been a scarcity of software packages that fast freely accessible. We have created an R package called dynr can handle broad class discrete- continuous-time models,...
Recent years have seen the emergence of an "idio-thetic" class methods to bridge gap between nomothetic and idiographic inference. These describe trends in processes by pooling intraindividual information across individuals inform group-level inference or vice versa. The current work introduces a novel model: subgrouped chain graphical vector autoregression (scGVAR). scGVAR is unique its ability identify subgroups who share common dynamic network structures both lag(1) contemporaneous...
In the past several decades, methodologies used to estimate nonlinear relationships among latent variables have been developed almost exclusively fit cross-sectional models. We present a relatively new estimation approach, unscented Kalman filter (UKF), and illustrate its potential as tool for fitting dynamic models in two ways: (1) building block approximating log–likelihood of state–space (2) time-varying wherein parameters are represented estimated online other variables. Furthermore,...
Dynamic factor analysis models with time-varying parameters offer a valuable tool for evaluating multivariate time series data dynamics and/or measurement properties. We use the Model of Activation proposed by Zautra and colleagues (Zautra, Potter, & Reich, 1997) as motivating example to construct dynamic model vector autoregressive relations cross-regression at level. Using techniques drawn from state-space literature, was fitted set daily affect (over 71 days) 10 participants who had been...
Myriad approaches for handling missing data exist in the literature. However, few studies have investigated tenability and utility of these when used with intensive longitudinal data. In this study, we compare illustrate two multiple imputation (MI) coping missingness fitting multivariate time-series models under different mechanisms. They include a full MI approach, which all dependent variables covariates are imputed simultaneously, partial MI, whereas is handled via information maximum...
This article models the chain of risk that links life course socioeconomic status (SES), daily stressor exposure and severity, well-being. Data from main survey diary project Midlife in United States (MIDUS) Refresher study were combined, resulting 782 participants (55.6% female; age 25–74, Mage = 47.9) who reported on 5,849 days information stressors measured at both person day levels. Between-person predictor variables include childhood SES, education, adult SES. Within-person assessed to...
Catastrophe theory (Thom, 1972(Thom, , 1993) ) is the study of many ways in which continuous changes a system's parameters can result discontinuous 1 or several outcome variables interest.Catastrophe theory-inspired models have been used to represent variety change phenomena realm social and behavioral sciences.Despite their promise, widespread applications catastrophe impeded, part, by difficulties performing model fitting comparison procedures.We propose new modeling framework for testing...
We compare the performances of well-known frequentist model fit indices (MFIs) and several Bayesian selection criteria (MCC) as tools for cross-loading in factor analysis under low to moderate sample sizes, possible violations distributional assumptions. The considered include Bayes (BF), Information Criterion (BIC), Deviance (DIC), a leave-one-out with Pareto smoothed importance sampling (LOO-PSIS), variable method using spike-and-slab prior (SSP; Lu, Chow, & Loken, 2016). Simulation...
The influx of intensive longitudinal data creates a pressing need for complex modeling tools that help enrich our understanding how individuals change over time. Multilevel vector autoregressive (mlVAR) models allow simultaneous evaluations reciprocal linkages between dynamic processes and individual differences, have gained increased recognition in recent years. High-dimensional other variations mlVAR models, though often computationally intractable the frequentist framework, can be readily...
Goal and aimsOne challenge using wearable sensors is nonwear time. Without a (e.g., capacitive) sensor, actigraphy data quality can be biased by subjective determinations confounding sleep/wake classification. We developed evaluated machine learning algorithm supplemented dynamic features to discern wear/nonwear episodes.Focus technologyActigraphy from wrist actigraph (Spectrum, Philips-Respironics).Reference technologyThe built-in sensor as "ground truth" classify periods other data,...
Background The onset of the COVID-19 pandemic in early 2020 introduced unprecedented disruptions impacting emotional well-being and daily routines US youths. However, patterns persistence these impacts over pandemic’s multiyear course remain less well understood. Objective This study examined longitudinal changes affect mobility observed adolescence young adulthood from June 2016 to April 2022. aimed quantify youths’ mood following response local case rates as effects course. Methods...
Background Regulating gestational weight gain (GWG) in pregnant women with overweight or obesity is difficult, particularly because of the narrow range recommended GWG for optimal health outcomes. Given that many show excessive and considering lack a “gold standard” intervention to manage GWG, there timely need effective efficient approaches regulate GWG. We have enhanced Healthy Mom Zone (HMZ) 2.0 novel digital platform, automated dosage changes, personalized strategies our pilot study...
Dynamic Structural Equation Models (DSEMs) integrate multilevel modeling, time series analysis, and structural equation modeling within a Bayesian estimation framework, offering versatile tool for analyzing intensive longitudinal data (ILD). However, the impact of measurement structure misspecification in DSEMs, especially under varying reliability conditions model complexities, remains underexplored. Our Monte Carlo simulation revealed that omitting errors when present led to severe biases...
Parameters in time series and other dynamic models often show complex range restrictions their distributions may deviate substantially from multivariate normal or standard parametric distributions. We use the truncated Dirichlet process (DP) as a non-parametric prior for such parameters novel nonlinear Bayesian factor analysis model. This is equivalent to specifying distribution be mixture composed of an unknown number discrete point masses (or clusters). The stick-breaking blocked Gibbs...