Structural equation model trees are data-driven tools for finding variables that predict group differences in SEM parameters. SEM trees build upon the decision tree paradigm by growing tree structures that divide a data set recursively into homogeneous subsets. In past research, SEM trees have been estimated predominantly with the R package semtree. The original algorithm in the semtree package selects split variables among covariates by calculating a likelihood ratio for each possible split of each covariate. Obtaining these likelihood ratios is (...) computationally demanding. As a remedy, we propose to guide the construction of SEM trees by a family of score-based tests that have recently been popularized in psychometrics. These score-based tests monitor fluctuations in case-wise derivatives of the likelihood function to detect parameter differences between groups. Compared to the likelihood-ratio approach, score-based tests are computationally efficient because they do not require refitting the model for every possible split. In this paper, we introduce score-guided SEM trees, implement them in semtree, and evaluate their performance by means of a Monte Carlo simulation. (shrink)

The beneficial effects of physical exercise on physical health and cognitive functioning have been repeatedly shown. However, evidence of its effect on psychosocial functioning in healthy adults is still scarce or inconclusive. One limitation of many studies examining this link is their reliance on correlational approaches or specific subpopulations, such as clinical populations. The present study investigated the effects of a physical exercise intervention on key factors of psychosocial functioning, specifically well-being, stress, loneliness, and future time perspective. We used data (...) from healthy, previously sedentary older adults who participated in a 6-month at-home intervention, either engaging in aerobic exercise or as part of a control group who participated in foreign language-learning or reading of selected native-language literature. Before and after the intervention, comprehensive cardiovascular pulmonary testing and a psychosocial questionnaire were administered. The exercise group showed significantly increased fitness compared to the control group. Contrary to expectations, however, we did not find evidence for a beneficial effect of this fitness improvement on any of the four domains of psychosocial functioning we assessed. This may be due to pronounced stability of such psychological traits in older age, especially in older adults who show high levels of well-being initially. Alternatively, it may be that the well-documented beneficial effects of physical exercise on brain structure and function, as well as cognition differ markedly from beneficial effects on psychosocial functioning. While aerobic exercise may be the driving factor for the former, positive effects on the latter may only be invoked by other aspects of exercise, for example, experiences of mastery or a feeling of community. (shrink)

In this article, we extend the Bayesian nonparametric regression method Gaussian Process Regression to the analysis of longitudinal panel data. We call this new approach Gaussian Process Panel Modeling (GPPM). GPPM provides great flexibility because of the large number of models it can represent. It allows classical statistical inference as well as machine learning inspired predictive modeling. GPPM offers frequentist and Bayesian inference without the need to resort to Markov chain Monte Carlo-based approximations, which makes the approach exact and fast. (...) GPPMs are defined using the kernel-language, which can express many traditional modeling approaches for longitudinal data, such as linear structural equation models, multilevel models, or state-space models but also various commonly used machine learning approaches. As a result, GPPM is uniquely able to represent hybrid models combining traditional parametric longitudinal models and nonparametric machine learning models. In the present paper, we introduce GPPM and illustrate its utility through theoretical arguments as well as simulated and empirical data. (shrink)

In aging humans, aerobic exercise interventions have been found to be associated with more positive or less negative changes in frontal and temporal brain areas, such as the anterior cingulate cortex (ACC) and hippocampus, relative to no-exercise control conditions. However, individual measures such as gray-matter (GM) probability may afford less reliable and valid conclusions about maintenance or losses in structural brain integrity than a latent construct based on multiple indicators. Here, we established a latent factor of GM structural integrity based (...) on GM probability assessed by voxel-based morphometry, magnetization transfer saturation, and mean diffusivity. Based on this latent factor, we investigated changes in structural brain integrity during a six-month exercise intervention in brain regions previously reported in studies using volumetric approaches. Seventy-five healthy, previously sedentary older adults aged 63–76 years completed an at-home intervention study in either an exercise group (EG;n= 40) or in an active control group (ACG;n= 35). Measures of peak oxygen uptake (VO2peak) taken before and after the intervention revealed a time-by-group interaction, with positive average change in the EG and no reliable mean change in the ACG. Significant group differences in structural brain integrity changes were observed in the right and left ACC, right posterior cingulate cortex (PCC), and left juxtapositional lobule cortex (JLC). In all instances, average changes in the EG did not differ reliably from zero, whereas average changes in the ACG were negative, pointing to maintenance of structural brain integrity in the EG, and to losses in the ACG. Significant individual differences in change were observed for right ACC and left JLC. Following up on these differences, we found that exercising participants with greater fitness gains also showed more positive changes in structural integrity. We discuss the benefits and limitations of a latent-factor approach to changes in structural brain integrity, and conclude that aerobic fitness interventions are likely to contribute to brain maintenance in old age. (shrink)