Dynamic Structural Equation Modeling Research Articles

Dynamic structural equation models synthesize ecosystem dynamics constrained by ecological mechanisms

Abstract Ecological analyses typically involve many interacting variables. Ecologists often specify lagged interactions in community dynamics (i.e. vector‐autoregressive models) or simultaneous interactions (e.g. structural equation models), but there is less familiarity with dynamic structural equation models (DSEM) that can include any simultaneous or lagged effect in multivariate time‐series analysis. We propose a novel approach to parameter estimation for DSEM, which involves constructing a Gaussian Markov random field (GMRF) representing simultaneous and lagged path coefficients, and then fitting this as a generalized linear mixed model to missing and/or non‐normal data. We provide a new R‐package dsem, which extends the ‘arrow interface’ from path analysis to represent user‐specified lags when constructing the GMRF. We also outline how the resulting nonseparable precision matrix can generalize existing separable models, for example, for time‐series and species interactions in a vector‐autoregressive model. We first demonstrate dsem by simulating a two‐species vector‐autoregressive model based on wolf–moose interactions on Isle Royale. We show that DSEM has improved precision when data are missing relative to a conventional dynamic linear model. We then demonstrate DSEM via two contrasting case studies. The first identifies a trophic cascade where decreased sunflower starfish has increased urchin and decreased kelp densities, while sea otters have a simultaneous positive effect on kelp in the California Current from 1999 to 2018. The second estimates how declining sea ice has decreased cold‐water habitats, driving a decreased density for fall copepod predation and inhibiting early‐life survival for Alaska pollock from 1963 to 2023. We conclude that DSEM can be fitted efficiently as a GLMM involving missing data, while allowing users to specify both simultaneous and lagged effects in a time‐series structural model. DSEM then allows conceptual models (developed with stakeholder input or from ecological expertise) to be fitted to incomplete time series and provides a simple interface for granular control over the number of estimated time‐series parameters. Finally, computational methods are sufficiently simple that DSEM can be embedded as component within larger (e.g. integrated population) models. We therefore recommend greater exploration and performance testing for DSEM relative to familiar time‐series forecasting methods.

The interdependence of substance use, satisfaction with life, and psychological distress: a dynamic structural equation model analysis.

Longitudinal studies with annual follow-up including psychological and social variables in substance use disorder recovery are scarce. We investigated whether levels of substance use, satisfaction with life, and psychological distress fluctuate across five years in relation to having drug-free friends. A prospective naturalistic cohort study of change trajectories in a cohort of people diagnosed with substance use disorder and using multiple substances with quarterly and annual follow-up over five years. Two-hundred-and-eight patients were recruited from substance use disorder treatment in Rogaland, Norway. Out of these, 164 participants fulfilled the inclusion criteria. We used Bayesian two-level dynamic structural equation modelling. The variable 'drug-free friends' was assessed by a self-reporting questionnaire, while psychological distress was assessed using the Symptoms Checklist 90 Revised. Satisfaction with life was assessed using the Satisfaction With Life Scale while drug use was assessed using the Drug Use Disorders Identification Test. The main findings are that higher-than-average psychological distress at a three-month lag credibly predicts higher-than-normal substance use at the concurrent time point t. Substance use and satisfaction with life seem to have synchronous trajectories over time, i.e. as the first decreases the latter increases and vice versa. During the five years after treatment, the participants mainly experienced a decrease in substance use and increase in satisfaction with life. Since the participants experienced positive and negative fluctuations for several years after treatment, it seems crucial to establish a dialogue with treatment professionals in order to create functional solutions for maintaining motivation and aiding recovery.

Modeling momentary reciprocal associations between negative affect and craving for alcohol and cannabis using dynamic structural equation modeling.

Modeling momentary reciprocal associations between negative affect and craving for alcohol and cannabis using dynamic structural equation modeling.

Longitudinal dynamics between anxiety and depression in bipolar spectrum disorders.

Anxiety and depression are common among individuals with bipolar spectrum disorders (BSDs), with anxiety being a risk factor for depression and vice versa. While the harmful effects of these symptoms are well recognized, their temporal dynamics have not been fully tested. To address this gap, our study investigated bidirectional relationships between anxiety and depression in individuals with BSDs using data from the Prechter Longitudinal Study of Bipolar Disorder, collected over an average of 11 years. We included 651 participants with various BSD subtypes (BD I, BD II, BD not otherwise specified, and schizoaffective bipolar type), with at least 5 years' data for adequate statistical power in detecting temporal dynamics. Bimonthly measurements of anxiety and depression were analyzed using dynamic structural equation modeling. Beyond assessing autoregressive and cross-lagged effects, this study also investigated whether temporal dynamics differed based on demographic characteristics and the use of psychiatric medication. Our findings revealed that individuals with BSDs experienced significant fluctuations in anxiety and depression over time. In addition, we found significant autoregressive and cross-lagged effects of anxiety and depression. Comparison of the cross-lagged effects demonstrated that anxiety had a greater effect on subsequent depression than vice versa. Age and marital status impacted cross-lagged and autoregressive effects. Specifically, older participants had stronger temporal associations between depression and subsequent anxiety, while widowed participants exhibited a heightened impact of depression on subsequent depression. These results underscore the importance of early identification and integrative interventions aimed at addressing both anxiety and depression to mitigate subsequent symptoms in BSDs. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Responsiveness, Social Connection, Hope, and Life Satisfaction in Everyday Social Interaction: An Experience Sampling Study

What are the specific everyday communication experiences—from across people’s social networks—that contribute to well-being? In the present work, we focus on the effects of perceived partner responsiveness in social interactions on various well-being outcomes. We hypothesized that everyday moments of responsiveness indirectly support two key estimates of well-being (hope and life satisfaction) through feelings of social connection. Data were obtained in an experience sampling study collected across ten days (N = 120). Results of dynamic structural equation modeling (DSEM) showed that responsive interaction predicted increases in hope (but not life satisfaction) through social connection. Results also identified reciprocal within-person links between responsive interaction and social connection throughout the day. These findings underscore the importance of responsive everyday communication for fostering social connection across different types of relationships and for supporting people’s capacity for a hopeful life. We discuss the implications of these results for continued research of responsiveness, hope theory, and well-being from a social interaction lens. On a practical level, the mediation pathway involving hope suggests how small changes in our patterns of everyday social interaction can be consequential to the quality of our lives.

Mediation Analyses of Intensive Longitudinal Data with Dynamic Structural Equation Modeling

Currently, dynamic structural equation modeling (DSEM) and residual DSEM (RDSEM) are commonly used in testing intensive longitudinal data (ILD). Researchers are interested in ILD mediation models, but their analyses are challenging. The present paper mathematically derived, empirically compared, and step-by-step demonstrated three types (i.e., 1-1-1, 2-1-1, and 2-2-1) of intensive longitudinal mediation (ILM) analyses based on DSEM and RDSEM models. Specifically, each ILM model was demonstrated with a simulated example and illustrated with the corresponding annotated Mplus codes. We compared two types of detrending methods in mediation analyses and showed that RDSEM was superior to DSEM because the latter included the time tj variable as a Level 1 predictor. Lastly, we extended ILM analyses based on DSEM and RDSEM to multilevel autoregressive mediation models, cross-classified DSEM, and intensive longitudinal moderated mediation models.

On liking how you look and feeling sexy: Quantifying the overlap between sexual desirability and body satisfaction in a sample of Australian Grindr users

The body image field aims to cultivate positive body image. To do so, it must appreciate factors contributing to positive body image. Sexual desirability is one such factor. Using experience sampling data from Australian Grindr users, we aimed to quantify the overlap between sexual desirability and body satisfaction. We found sexual desirability and body satisfaction correlated strongly between- (r = .90) and within-persons (rMedian =.60). Using dynamic structural equation modeling, we analyzed 238 participants' data (T = 9058), finding that sexual desirability and body satisfaction were bidirectionally related – previous sexual desirability predicted current body satisfaction (β = 0.22) and vice versa (β = 0.17). Participants’ average body satisfaction tended to be higher when sexual desirability contributed more to body satisfaction (r = .31) and was more stable across time (r = .19). We found sexual desirability and body satisfaction overlap considerably and that sexual desirability may contribute more to body satisfaction than vice versa. Our results suggest that (1) sexual desirability and its dynamics across time influence body satisfaction, and (2) research on sexual desirability and its relationship with body image should be a focus for the field. We discuss avenues for future research on sexual desirability.

Stress and emotion in a locked campus: the moderating effects of resilience and loneliness.

The aim of this study is to investigate the dynamic relationship between Chinese students' emotions and stress during a strict lockdown period in a university setting and the context of a global pandemic. Dynamic structural equation modeling was used to investigate the moderating role of resilience and loneliness in this relationship. The participants consisted of 112 students. Based on loneliness and resilience measures and the intensive tracking of emotional stress over a 21-day period, the results of data analysis indicated that the students' overall levels of positive emotions were low and relatively independent of negative emotions. Negative emotions were significantly autoregressive and their baseline was closely related to the individual's overall feelings of stress and loneliness levels, fluctuating with feelings of stress. The results confirm the hypothesis that resilience helps to stabilize emotions. Individuals with low resilience may be more emotionally sensitive in confined environments, while receiving social support may help to alleviate low moods.

Dynamic Structural Equation Models: Promising Yet Concerning

Dynamic Structural Equation Model (DSEM) is a powerful statistical modeling approach that has recently gained popularity among researchers studying intensive longitudinal data. Despite its exciting potential, the stability and replicability of DSEM is yet to be closely examined. This study empirically investigates DSEM using recently published data to explore its strengths and potential limitations. The results show that while some of its parameter estimates are stable, others are characterized by substantial variation as a function of seemingly innocuous initial model estimation conditions. Indeed, some parameters fluctuate between significance and non-significance for the same model estimated using the same data. The instability of DSEM estimates poses a serious threat to the internal and external validity of conclusions drawn from its analyses, challenging the reproducibility of findings from applied research. Given the recent focus on the replication crisis in psychology, it is critical to address these issues as the popularity of DSEM in psychological research continues to rise. Several potential solutions are investigated to address this problem and recommendations of best practice are offered to applied researchers who plan to use DSEM in intensive longitudinal data analysis. KEYWORDS: Dynamic Structural Equation Model; Bayesian; Robust Estimation; Intensive Longitudinal Data

From cross-lagged effects to feedback effects: Further insights into the estimation and interpretation of bidirectional relations.

Bidirectional relations have long been of interest in psychology and other social behavioral sciences. In recent years, the widespread use of intensive longitudinal data has provided new opportunities to examine dynamic bidirectional relations between variables. However, most previous studies have focused on the effect of one variable on the other (i.e., cross-lagged effects) rather than the overall effect representing the dynamic interplay between two variables (i.e., feedback effects), which we believe may be due to a lack of relevant methodological guidance. To quantify bidirectional relations as a whole, this study attempted to provide guidance for the estimation and interpretation of feedback effects based on dynamic structural equation models. First, we illustrated the estimation procedure for the average and person-specific feedback effects. Then, to facilitate the interpretation of feedback effects, we established an empirical benchmark by quantitatively synthesizing the results of relevant empirical studies. Finally, we used a set of empirical data to demonstrate how feedback effects can help (a) test theories based on bidirectional relations and (b) reveal correlates of individual differences in bidirectional relations. We also discussed the broad application prospects of feedback effects from a dynamic systems perspective. This study provides guidance for applied researchers interested in further examining feedback effects in bidirectional relations, and the shift from focusing on cross-lagged effects only to a comprehensive consideration of feedback effects may provide new insights into the study of bidirectional relations.

DYNAMIC IMPACTS OF FAMILY CONFLICT ON ELDER ABUSE AND NEGLECT AMONG ADRD CAREGIVERS

Abstract Conflict with other family members impacts dementia caregivers’ well-being and stress-appraisals and may also influence their use of elder abuse and neglect (EAN). To test this, we utilized latent class analysis to classify spouse, adult child, grandchild, and nontraditional (step- and in-laws) caregivers (N = 453) based on family conflict levels at baseline. Caregivers were classified into four groups: “low,” “moderate,” “high,” and “EAN accusations.” Spouses were significantly more likely to be in the “low” class compared to other classes. Nontraditional caregivers were more likely to be in the “EAN accusations” class versus the “moderate” class (β = 1.73, SE = .70, p < .05). Belonging to the EAN accusations class predicted EAN at baseline (t = 3.728, p < .001). Next, dynamic structural equation modeling was used to test whether stress from family conflict on a given day predicts use of EAN behaviors. Family conflict stress predicted EAN on the same day (β = 0.05, 95% CI = 0.03-0.07), but not the subsequent day (CL= -0.01, 95% CI = 0.03-0.02). Similar models were conducted for each relationship type. The effect of family conflict stress on EAN was not significant for spouse and grandchildren but remained significant for adult child and non-traditional caregivers. Findings suggest interventions related to conflict management and communication among caregiving families may be beneficial new intervention targets for reducing EAN in ADRD. Addressing family conflict may be most important when the caregiver is an adult child or nontraditional relationship.

Purpose and goal pursuit as a self‐sustaining system: Evidence of daily within‐person reciprocity among adolescents in self‐driven learning

AbstractObjectiveDespite long‐standing assumptions that a sense of purpose in life and goal pursuit are mutually supportive, empirical evidence of their reciprocity remains deficient. In the context of a unique out‐of‐school time program that empowers youth to pursue passions through self‐driven learning, we examined whether purpose and one aspect of goal pursuit—perceptions of goal progress—work together to sustain themselves and each other over time.MethodAdolescents (N = 321) completed daily surveys throughout program enrollment (Menrollment = 69.09 days). Through dynamic structural equation modeling, we derived within‐person patterns of day‐to‐day prediction as well as individual differences in these patterns.ResultsWe found purpose and perceived goal progress exhibited significant daily inertia (i.e., autoregressive prediction) and reciprocity (i.e., cross‐lagged prediction) at the within‐person level. We also found initial evidence suggesting (a) tighter reciprocity was related to greater perceived goal progress overall and (b) people with greater purpose inertia may rely less on making goal progress to sustain momentum.ConclusionsWith evidence of daily purpose‐progress reciprocity, the field can look forward to replicating this work in other contexts, diving deeper into interesting patterns of within‐person dynamics, and developing interventions to support youth striving.

Using learning analytics with temporal modeling to uncover the interplay of before-class video viewing engagement, motivation, and performance in an active learning context

Video learning is a crucial component in flipped learning related to learners' readiness to attend more complex face-to-face activities. However, instructors may not know students' preparedness and motivation for each class. This study intended to examine the underlying mechanism between students' before-class video-viewing engagement and performance. We investigated learners' motivation (i.e., autonomous and controlled motivation) based on self-determination theory and controlled for their gender, age, and prior performance. An integrated temporal model was built under the dynamic structural equation modeling framework. Participants were 118 Taiwanese graduate students (78 females and 40 males) with a mean age of 26.14 years old. Study findings showed that students engage with the video content in a consistent viewing pattern, which has a significant relationship with their motivation and academic performance from week to week. The frequency of skipping backward was negatively associated with controlled motivation, while the frequency of pauses was positively associated with autonomous and controlled motivation. A high frequency of skipping forward predicted lower learning performance. Besides, a higher ratio of skipping backward predicted higher controlled motivation. Male students, older students, and students with higher prior performance exhibited higher autonomous motivation, which was related to better learning performance. Understanding these relationships can help educators and instructional designers develop more effective strategies for fostering students’ preparedness, engagement and success in active learning contexts. The study findings were discussed considering theoretical advancement and practical implications for enhancing personalized and active learning in higher education to achieve educational equity.

Control-value appraisals and academic emotions: An intensive longitudinal examination of reciprocal effects.

This study examined the reciprocal relation between lesson-specific perceived cognitive appraisals and academic emotions on an intra-individual level. A daily diary study was conducted using a sample of 266 Chinese Han students (Grades 7-8; 56.8% boys; Mage = 13.70, SDage = 0.52) during 10 mathematics lessons in 2022. Standardized questionnaires were also administered to these students before the daily diary study. The results of the dynamic structural equation modeling revealed significant reciprocal relations between cognitive appraisals and academic emotions within early adolescents and highlighted the role of emotions in guiding cognitive appraisals. Additionally, the study identified similarities and differences in the inter-individual relation between appraisals and emotions across self-reported questionnaires and daily diary measures.

Examining Environmental Predictors of Behavioral Symptoms of Dementia Using Dynamic Structural Equation Modelling

AbstractBackgroundMost persons with Alzheimer’s disease and related dementias (ADRD) experience behavioral symptoms as a result of exposure to stressors in the environment. Daily diaries can capture fluctuations in behavioral symptoms in response to environmental stressors. The purpose of this multi‐time series longitudinal study was to determine whether potential protective factors, caregiver reciprocity and a shared activity were associated with the number of behavioral symptoms person with ADRD.MethodA community‐dwelling sample of 453 caregivers for persons with ADRD (age 18+) completed 21 days of once‐a‐day diary surveys. Caregiver perceived reciprocity was assessed using a question asking, “How satisfied were you with amount of affection or appreciation your relative with dementia showed towards you?” Responses were coded using a Likert scale with 1‐Not at all and 5‐Very much. Caregivers were asked whether they did a pleasant non‐care activity with their relative and about a presence of eight different behavioral symptoms in persons with ADRD. We used a two‐level dynamic structural equation model to examine the relationship between the number of different behavioral symptoms, caregiver reciprocity and a shared activity on a given day, and the following day.ResultCaregivers completed 7,783 diaries. Caregivers were mostly female (87.4%), non‐Hispanic white (51.4%) with a mean age of 53 years (SD = 14) caring for a parent/parent‐in‐law/stepparent (61.5%) with a mean age of 79 years (SD = 9). Caregiver perceived reciprocity had a significant direct effect on the number of behavioral symptoms experienced the same day and the next day. While no significant main or cross‐lagged effects were found between shared activity and behavioral symptoms, estimates suggest the impact was negligible (0.7% increase).ConclusionWe found that caregiver perceived reciprocity may reduce the number of behavioral symptoms a person with ADRD has that day and the following day. These findings support the growing body of theoretical and empirical work suggesting caregivers are an important feature of the environment and influence behavioral symptom presentation.

Novel machine learning approaches for improving the reproducibility and reliability of functional and effective connectivity from functional MRI

Objective. New measures of human brain connectivity are needed to address gaps in the existing measures and facilitate the study of brain function, cognitive capacity, and identify early markers of human disease. Traditional approaches to measure functional connectivity (FC) between pairs of brain regions in functional MRI, such as correlation and partial correlation, fail to capture nonlinear aspects in the regional associations. We propose a new machine learning based measure of FC ( ML.FC ) which efficiently captures linear and nonlinear aspects. Approach. To capture directed information flow between brain regions, effective connectivity (EC) metrics, including dynamic causal modeling and structural equation modeling have been used. However, these methods are impractical to compute across the many regions of the whole brain. Therefore, we propose two new EC measures. The first, a machine learning based measure of effective connectivity ( ML.EC ), measures nonlinear aspects across the entire brain. The second, Structurally Projected Granger Causality ( SP.GC ) adapts Granger Causal connectivity to efficiently characterize and regularize the whole brain EC connectome to respect underlying biological structural connectivity. The proposed measures are compared to traditional measures in terms of reproducibility and the ability to predict individual traits in order to demonstrate these measures’ internal validity. We use four repeat scans of the same individuals from the Human Connectome Project and measure the ability of the measures to predict individual subject physiologic and cognitive traits. Main results. The proposed new FC measure of ML.FC attains high reproducibility (mean intra-subject R 2 of 0.44), while the proposed EC measure of SP.GC attains the highest predictive power (mean R 2 across prediction tasks of 0.66). Significance. The proposed methods are highly suitable for achieving high reproducibility and predictiveness and demonstrate their strong potential for future neuroimaging studies.

Long-term aging trajectories of the accumulation of disease burden as predictors of daily affect dynamics and stressor reactivity.

Multiple-timescale studies provide new opportunities to examine how developmental processes that evolve at different cadences are intertwined. Developmental theories of emotion regulation suggest that the long-term, slowly evolving age-related accumulation of disease burden should shape short-term, faster evolving (daily) affective experiences. To empirically examine this proposition, we combined data from 123 old adults (65-69 years, 47% women) and 32 very old adults (85-88 years, 59% women) who provided 20 + year within-person longitudinal data on physician-rated morbidity and subsequently also completed repeated daily-life assessments of stress and affect six times a day over 7 consecutive days as they were going about their daily-life routines. Results from models that simultaneously articulate growth and intraindividual variability processes (in a dynamic structural equation modeling framework) revealed that individual differences in long-term aging trajectories of the accumulation of disease burden were indeed predictive of differences in three facets of affective dynamics that manifest in everyday life. In particular-over and above mean levels of disease burden-older adults whose disease burden had increased more over the past 20 years had higher base level of negative affect in their daily lives, more emotional reactivity to the experience of daily stressors, and more moment-to-moment fluctuations in negative affect that was unrelated to stressors (affective systemic noise). We highlight that developmental processes evolving over vastly different timescales are intertwined, and speculate how new knowledge about those relations can inform developmental theories of emotion regulation and daily-life functioning. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Investigating Moderation Effects at the Within-Person Level Using Intensive Longitudinal Data: A Two-Level Dynamic Structural Equation Modelling Approach in Mplus

Recent technological advances have provided new opportunities for the collection of intensive longitudinal data. Using methods such as dynamic structural equation modeling, these data can provide new insights into moment-to-moment dynamics of psychological and behavioral processes. In intensive longitudinal data (t > 20), researchers often have theories that imply that factors that change from moment to moment within individuals act as moderators. For instance, a person’s level of sleep deprivation may affect how much an external stressor affects mood. Here, we describe how researchers can implement, test, and interpret dynamically changing within-person moderation effects using two-level dynamic structural equation modeling as implemented in the structural equation modeling software Mplus. We illustrate the analysis of within-person moderation effects using an empirical example investigating whether changes in spending time online using social media affect the moment-to-moment effect of loneliness on depressive symptoms, and highlight avenues for future methodological development. We provide annotated Mplus code, enabling researchers to better isolate, estimate, and interpret the complexities of within-person interaction effects.

I feel smart today! A daily diary study on narcissism and self-assessed intelligence

While the objective level of intelligence is not associated with narcissism, relations to self-assessed intelligence (SAI) have been repeatedly reported. Existing research suggests that different facets of narcissism may have different associations with SAI. In the current daily diary study ( N = 176; N = 3975 total observations), we employed dynamic structural equation modeling to examine the relationships between facets of trait and state narcissism (i.e., agentic, antagonistic, and neurotic) and the level, variability, and instability of SAI assessed over 28 consecutive days. Both trait and state narcissism were consistently related to SAI: agentic narcissism showed a positive relationship, whereas antagonistic and neurotic narcissism showed negative relationships with SAI. Trait agentic and state antagonistic narcissism predicted greater variability of SAI scores throughout the study, while neither trait nor state narcissism predicted the instability of SAI. Finally, we found that experiencing increased agentic narcissism on one day, predicted perceiving oneself as more intelligent on the next day, but feeling smarter did not predict feeling narcissistic over time. Moreover, we demonstrated that differentiating between narcissism facets yielded more theoretically accurate results compared to distinguishing between grandiose and vulnerable narcissism.

The daily dynamics of basic psychological need satisfaction at work, their determinants, and their implications: An application of Dynamic Structural Equation Modeling

ABSTRACT Drawing on self-determination theory, this study focuses on the person- and occasion-specific components of the daily dynamics of employees’ global psychological need satisfaction at work. Predictors (job demands related to information and communication technologies, segmentation norms, and workload) and outcomes (perceived productivity, psychological detachment, work-family conflict, job satisfaction, and personal satisfaction) were also examined across both levels to better grasp the mechanisms underlying these short-term dynamics. A total of 129 French employees filled out questionnaire surveys at the end of each workday for five days (521 observations). Results from Dynamic Structural Equation Modeling (DSEM) showed clear associations between need satisfaction, the predictors, and the outcomes at the person-specific level. However, and although need satisfaction levels were found to fluctuate on a daily basis, they seemed immune to the effects of daily fluctuations in predictor levels, and unlikely to generate matching fluctuations in outcome levels. These results suggest strong homoeostatic processes protecting employees’ functioning against daily fluctuations, but that the accumulation of such fluctuations over the work week may jeopardize these processes.