Network Inference With the Lasso.

Calculating confidence intervals and p-values of edges in networks is useful to decide their presence or absence and it is a natural way to quantify uncertainty. Since lasso estimation is often used to obtain edges in a network, and the underlying distribution of lasso estimates is discontinuous and has probability one at zero when the estimate is zero, obtaining p-values and confidence intervals is problematic. It is also not always desirable to use the lasso to select the edges because there are assumptions required for correct identification of network edges that may not be warranted for the data at hand. Here, we review three methods that either use a modified lasso estimate (desparsified or debiased lasso) or a method that uses the lasso for selection and then determines p-values without the lasso. We compare these three methods with popular methods to estimate Gaussian Graphical Models in simulations and conclude that the desparsified lasso and its bootstrapped version appear to be the best choices for selection and quantifying uncertainty with confidence intervals and p-values.

Life meaning and feelings of ineffectiveness as transdiagnostic factors in eating disorder and comorbid internalizing symptomatology - A combined undirected and causal network approach.

The field of eating disorders is facing problems ranging from a suboptimal classification system to low long-term success rates of treatments. There is evidence supporting a transdiagnostic approach to explain the development and maintenance of eating disorders. Meaning in life has been proposed as a promising key transdiagnostic factor that could potentially not only bridge between the different eating disorder subtypes but also explain frequent co-occurrence with symptoms of comorbid psychopathology, such as anxiety and depression. The present study used self-report data from 501 participants to construct networks of eating disorder and comorbid internalizing symptomatology, including factors related to meaning in life, i.e., presence of life meaning, perceived ineffectiveness, and satisfaction with basic psychological needs. In an undirected network model, it was found that ineffectiveness is a central node, also bridging between eating disorder and other psychological symptoms. A directed network model displayed evidence for a causal effect of presence of life meaning both on the core symptomatology of eating disorders and depressive symptoms via ineffectiveness. These results support the notion of meaning in life and feelings of ineffectiveness as transdiagnostic factors within eating disorder symptomatology in the general population.

Formalizing psychological interventions through network control theory.

Despite the growing deployment of network representation to comprehend psychological phenomena, the question of whether and how networks can effectively describe the effects of psychological interventions remains elusive. Network control theory, the engineering study of networked interventions, has recently emerged as a viable methodology to characterize and guide interventions. However, there is a scarcity of empirical studies testing the extent to which it can be useful within a psychological context. In this paper, we investigate a representative psychological intervention experiment, use network control theory to model the intervention and predict its effect. Using this data, we showed that: (1) the observed psychological effect, in terms of sensitivity and specificity, relates to the regional network control theoretic metrics (average and modal controllability), (2) the size of change following intervention negatively correlates with a whole-network topology that quantifies the "ease" of change as described by control theory (control energy), and (3) responses after intervention can be predicted based on formal results from control theory. These insights assert that network control theory has significant potential as a tool for investigating psychological interventions. Drawing on this specific example and the overarching framework of network control theory, we further elaborate on the conceptualization of psychological interventions, methodological considerations, and future directions in this burgeoning field.

How compliance with behavioural measures during the initial phase of a pandemic develops over time: A longitudinal COVID-19 study.

In this longitudinal research, we adopt a complexity approach to examine the temporal dynamics of variables related to compliance with behavioural measures during the COVID-19 pandemic. Dutch participants (N = 2399) completed surveys with COVID-19-related variables five times over a period of 10 weeks (23 April-30 June 2020). With these data, we estimated within-person COVID-19 attitude networks containing a broad set of psychological variables and their relations. These networks display variables' predictive effects over time between measurements and contemporaneous effects during measurements. Results show (1) bidirectional effects between multiple variables relevant for compliance, forming potential feedback loops, and (2) a positive reinforcing structure between compliance, support for behavioural measures, involvement in the pandemic and vaccination intention. These results can explain why levels of these variables decreased throughout the course of the study. The reinforcing structure points towards potentially amplifying effects of interventions on these variables and might inform processes of polarization. We conclude that adopting a complexity approach might contribute to understanding protective behaviour in the initial phase of pandemics by combining different theoretical models and modelling bidirectional effects between variables. Future research could build upon this research by studying causality with interventions and including additional variables in the networks.

The impact of ordinal scales on Gaussian mixture recovery.

Gaussian mixture models (GMMs) are a popular and versatile tool for exploring heterogeneity in multivariate continuous data. Arguably the most popular way to estimate GMMs is via the expectation-maximization (EM) algorithm combined with model selection using the Bayesian information criterion (BIC). If the GMM is correctly specified, this estimation procedure has been demonstrated to have high recovery performance. However, in many situations, the data are not continuous but ordinal, for example when assessing symptom severity in medical data or modeling the responses in a survey. For such situations, it is unknown how well the EM algorithm and the BIC perform in GMM recovery. In the present paper, we investigate this question by simulating data from various GMMs, thresholding them in ordinal categories and evaluating recovery performance. We show that the number of components can be estimated reliably if the number of ordinal categories and the number of variables is high enough. However, the estimates of the parameters of the component models are biased independent of sample size. Finally, we discuss alternative modeling approaches which might be adopted for the situations in which estimating a GMM is not acceptable.

Towards an encompassing theory of network models: Reply to Brusco, Steinley, Hoffman, Davis-Stober, and Wasserman (2019).

Network models like the Ising model are increasingly used in psychological research. In a recent article published in this journal, Brusco et al. (2019) provide a critical assessment of the conditions that underlie the Ising model and the eLasso method that is commonly used to estimate it. In this commentary, we show that their main criticisms are unfounded. First, where Brusco et al. (2019) suggest that Ising models have little to do with classical network models such as random graphs, we show that they can be fruitfully connected. Second, if one makes this connection it is immediately evident that Brusco et al.'s (2019) second criticism-that the Ising model requires complete population homogeneity and does not allow for individual differences in network structure-is incorrect. In particular, we establish that if every individual has their own topology, and these individual differences instantiate a random graph model, the Ising model will hold in the population. Hence, population homogeneity is sufficient for the Ising model, but it is not necessary, as Brusco et al. (2019) suggest. Third, we address Brusco et al.'s (2019) criticism regarding the sparsity assumption that is made in common uses of the Ising model. We show that this criticism is misdirected, as it targets a particular estimation algorithm for the Ising model rather than the model itself. We also describe various established and validated approaches for estimating the Ising model for networks that violate the sparsity assumption. Finally, we outline important avenues for future research. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Tailored interventions into broad attitude networks towards the COVID-19 pandemic.

This study examines how broad attitude networks are affected by tailored interventions aimed at variables selected based on their connectiveness with other variables. We first computed a broad attitude network based on a large-scale cross-sectional COVID-19 survey (N = 6,093). Over a period of approximately 10 weeks, participants were invited five times to complete this survey, with the third and fifth wave including interventions aimed at manipulating specific variables in the broad COVID-19 attitude network. Results suggest that targeted interventions that yield relatively strong effects on variables central to a broad attitude network have downstream effects on connected variables, which can be partially explained by the variables the interventions were aimed at. We conclude that broad attitude network structures can reveal important relations between variables that can help to design new interventions.

Comparing network structures on three aspects: A permutation test.

Network approaches to psychometric constructs, in which constructs are modeled in terms of interactions between their constituent factors, have rapidly gained popularity in psychology. Applications of such network approaches to various psychological constructs have recently moved from a descriptive stance, in which the goal is to estimate the network structure that pertains to a construct, to a more comparative stance, in which the goal is to compare network structures across populations. However, the statistical tools to do so are lacking. In this article, we present the network comparison test (NCT), which uses resampling-based permutation testing to compare network structures from two independent, cross-sectional data sets on invariance of (a) network structure, (b) edge (connection) strength, and (c) global strength. Performance of NCT is evaluated in simulations that show NCT to perform well in various circumstances for all three tests: The Type I error rate is close to the nominal significance level, and power proves sufficiently high if sample size and difference between networks are substantial. We illustrate NCT by comparing depression symptom networks of males and females. Possible extensions of NCT are discussed. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Modeling psychopathology: From data models to formal theories.

Over the past decade, there has been a surge of empirical research investigating mental disorders as complex systems. In this article, we investigate how to best make use of this growing body of empirical research and move the field toward its fundamental aims of explaining, predicting, and controlling psychopathology. We first review the contemporary philosophy of science literature on scientific theories and argue that fully achieving the aims of explanation, prediction, and control requires that we construct formal theories of mental disorders: theories expressed in the language of mathematics or a computational programming language. We then investigate three routes by which one can use empirical findings (i.e., data models) to construct formal theories: (a) using data models themselves as formal theories, (b) using data models to infer formal theories, and (c) comparing empirical data models to theory-implied data models in order to evaluate and refine an existing formal theory. We argue that the third approach is the most promising path forward. We conclude by introducing the abductive formal theory construction (AFTC) framework, informed by both our review of philosophy of science and our methodological investigation. We argue that this approach provides a clear and promising way forward for using empirical research to inform the generation, development, and testing of formal theories both in the domain of psychopathology and in the broader field of psychological science. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Relations between Networks, Regression, Partial Correlation, and the Latent Variable Model.

The Gaussian graphical model (GGM) has become a popular tool for analyzing networks of psychological variables. In a recent article in this journal, Forbes, Wright, Markon, and Krueger (FWMK) voiced the concern that GGMs that are estimated from partial correlations wrongfully remove the variance that is shared by its constituents. If true, this concern has grave consequences for the application of GGMs. Indeed, if partial correlations only capture the unique covariances, then the data that come from a unidimensional latent variable model ULVM should be associated with an empty network (no edges), as there are no unique covariances in a ULVM. We know that this cannot be true, which suggests that FWMK are missing something with their claim. We introduce a connection between the ULVM and the GGM and use that connection to prove that we find a fully-connected and not an empty network associated with a ULVM. We then use the relation between GGMs and linear regression to show that the partial correlation indeed does not remove the common variance.

The search for causality: A comparison of different techniques for causal inference graphs.

Estimating causal relations between two or more variables is an important topic in psychology. Establishing a causal relation between two variables can help us in answering that question of why something happens. However, using solely observational data are insufficient to get the complete causal picture. The combination of observational and experimental data may give adequate information to properly estimate causal relations. In this study, we consider the conditions where estimating causal relations might work and we show how well different algorithms, namely the Peter and Clark algorithm, the Downward Ranking of Feed-Forward Loops algorithm, the Transitive Reduction for Weighted Signed Digraphs algorithm, the Invariant Causal Prediction (ICP) algorithm and the Hidden Invariant Causal Prediction (HICP) algorithm, determine causal relations in a simulation study. Results showed that the ICP and the HICP algorithms perform best in most simulation conditions. We also apply every algorithm to an empirical example to show the similarities and differences between the algorithms. We believe that the combination of the ICP and the HICP algorithm may be suitable to be used in future research. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Invisible Hands and Fine Calipers: A Call to Use Formal Theory as a Toolkit for Theory Construction.

In recent years, a growing chorus of researchers has argued that psychological theory is in a state of crisis: Theories are rarely developed in a way that indicates an accumulation of knowledge. Paul Meehl raised this very concern more than 40 years ago. Yet in the ensuing decades, little has improved. We aim to chart a better path forward for psychological theory by revisiting Meehl's criticisms, his proposed solution, and the reasons his solution failed to meaningfully change the status of psychological theory. We argue that Meehl identified serious shortcomings in our evaluation of psychological theories and that his proposed solution would substantially strengthen theory testing. However, we also argue that Meehl failed to provide researchers with the tools necessary to construct the kinds of rigorous theories his approach required. To advance psychological theory, we must equip researchers with tools that allow them to better generate, evaluate, and develop their theories. We argue that formal theories provide this much-needed set of tools, equipping researchers with tools for thinking, evaluating explanation, enhancing measurement, informing theory development, and promoting the collaborative construction of psychological theories.

Moderated Network Models.

Pairwise network models such as the Gaussian Graphical Model (GGM) are a powerful and intuitive way to analyze dependencies in multivariate data. A key assumption of the GGM is that each pairwise interaction is independent of the values of all other variables. However, in psychological research, this is often implausible. In this article, we extend the GGM by allowing each pairwise interaction between two variables to be moderated by (a subset of) all other variables in the model, and thereby introduce a Moderated Network Model (MNM). We show how to construct MNMs and propose an ℓ1-regularized nodewise regression approach to estimate them. We provide performance results in a simulation study and show that MNMs outperform the split-sample based methods Network Comparison Test (NCT) and Fused Graphical Lasso (FGL) in detecting moderation effects. Finally, we provide a fully reproducible tutorial on how to estimate MNMs with the R-package mgm and discuss possible issues with model misspecification.

Latent Variable Models and Networks: Statistical Equivalence and Testability.

Networks are gaining popularity as an alternative to latent variable models for representing psychological constructs. Whereas latent variable approaches introduce unobserved common causes to explain the relations among observed variables, network approaches posit direct causal relations between observed variables. While these approaches lead to radically different understandings of the psychological constructs of interest, recent articles have established mathematical equivalences that hold between network models and latent variable models. We argue that the fact that for any model from one class there is an equivalent model from the other class does not mean that both models are equally plausible accounts of the data-generating mechanism. In many cases the constraints that are meaningful in one framework translate to constraints in the equivalent model that lack a clear interpretation in the other framework. Finally, we discuss three diverging predictions for the relation between zero-order correlations and partial correlations implied by sparse network models and unidimensional factor models. We propose a test procedure that compares the likelihoods of these models in light of these diverging implications. We use an empirical example to illustrate our argument.

Interpreting the Ising Model: The Input Matters.

The Ising model is a model for pairwise interactions between binary variables that has become popular in the psychological sciences. It has been first introduced as a theoretical model for the alignment between positive (1) and negative (-1) atom spins. In many psychological applications, however, the Ising model is defined on the domain {0, 1} instead of the classical domain {-1,1}. While it is possible to transform the parameters of the Ising model in one domain to obtain a statistically equivalent model in the other domain, the parameters in the two versions of the Ising model lend themselves to different interpretations and imply different dynamics, when studying the Ising model as a dynamical system. In this tutorial paper, we provide an accessible discussion of the interpretation of threshold and interaction parameters in the two domains and show how the dynamics of the Ising model depends on the choice of domain. Finally, we provide a transformation that allows one to transform the parameters in an Ising model in one domain into a statistically equivalent Ising model in the other domain.

A Tutorial on Estimating Time-Varying Vector Autoregressive Models.

Time series of individual subjects have become a common data type in psychological research. These data allow one to estimate models of within-subject dynamics, and thereby avoid the notorious problem of making within-subjects inferences from between-subjects data, and naturally address heterogeneity between subjects. A popular model for these data is the Vector Autoregressive (VAR) model, in which each variable is predicted by a linear function of all variables at previous time points. A key assumption of this model is that its parameters are constant (or stationary) across time. However, in many areas of psychological research time-varying parameters are plausible or even the subject of study. In this tutorial paper, we introduce methods to estimate time-varying VAR models based on splines and kernel-smoothing with/without regularization. We use simulations to evaluate the relative performance of all methods in scenarios typical in applied research, and discuss their strengths and weaknesses. Finally, we provide a step-by-step tutorial showing how to apply the discussed methods to an openly available time series of mood-related measurements.

Borderline personality disorder classification based on brain network measures during emotion regulation.

Borderline Personality Disorder (BPD) is characterized by an increased emotional sensitivity and dysfunctional capacity to regulate emotions. While amygdala and prefrontal cortex interactions are regarded as the critical neural mechanisms underlying these problems, the empirical evidence hereof is inconsistent. In the current study, we aimed to systematically test different properties of brain connectivity and evaluate the predictive power to detect borderline personality disorder. Patients with borderline personality disorder (n = 51), cluster C personality disorder (n = 26) and non-patient controls (n = 44), performed an fMRI emotion regulation task. Brain network analyses focused on two properties of task-related connectivity: phasic refers to task-event dependent changes in connectivity, while tonic was defined as task-stable background connectivity. Three different network measures were estimated (strength, local efficiency, and participation coefficient) and entered as separate models in a nested cross-validated linear support vector machine classification analysis. Borderline personality disorder vs. non-patient controls classification showed a balanced accuracy of 55%, which was not significant under a permutation null-model, p = 0.23. Exploratory analyses did indicate that the tonic strength model was the highest performing model (balanced accuracy 62%), and the amygdala was one of the most important features. Despite being one of the largest data-sets in the field of BPD fMRI research, the sample size may have been limited for this type of classification analysis. The results and analytic procedures do provide starting points for future research, focusing on network measures of tonic connectivity, and potentially focusing on subgroups of BPD.

Hippocampus plays a role in speech feedback processing.

There is increasing evidence that the hippocampus is involved in language production and verbal communication, although little is known about its possible role. According to one view, hippocampus contributes semantic memory to spoken language. Alternatively, hippocampus is involved in the processing the (mis)match between expected sensory consequences of speaking and the perceived speech feedback. In the current study, we re-analysed functional magnetic resonance (fMRI) data of two overt picture-naming studies to test whether hippocampus is involved in speech production and, if so, whether the results can distinguish between a "pure memory" versus a "prediction" account of hippocampal involvement. In both studies, participants overtly named pictures during scanning while hearing their own speech feedback unimpededly or impaired by a superimposed noise mask. Results showed decreased hippocampal activity when speech feedback was impaired, compared to when feedback was unimpeded. Further, we found increased functional coupling between auditory cortex and hippocampus during unimpeded speech feedback, compared to impaired feedback. Finally, we found significant functional coupling between a hippocampal/supplementary motor area (SMA) interaction term and auditory cortex, anterior cingulate cortex and cerebellum during overt picture naming, but not during listening to one's own pre-recorded voice. These findings indicate that hippocampus plays a role in speech production that is in accordance with a "prediction" view of hippocampal functioning.

Short-term sequences of aggressive behavior in psychiatric inpatients with psychotic disorders using Markov models.

BACKGROUND: Aggression in inpatients with psychotic disorders is harmful to patients and health care professionals. AIMS: The current study introduces a novel approach for assessing short-term sequences of different types of aggression. METHODS: Occurrence and type of aggressive behavior was assessed retrospectively by reviewing hospital charts in a sample of 120 inpatients with psychotic disorders, admitted to the psychiatric wards of an academic hospital using the Modified Overt Aggression Scale (MOAS). Behavioral sequences of verbal aggression, physical aggression against objects, physical aggression against oneself and physical aggression against others were analyzed by using Markov models, a statistical technique providing the probabilities of transferring from one state to another. RESULTS: The Markov models showed that when patients behave aggressively, they are likely to either show the same type of aggression or to be non-aggressive consecutively. Patients are, however, unlikely to subsequently show another type of aggression. Non-aggressive behavior is very unlikely to result in physical aggression or aggression against objects. CONCLUSION: The current study introduced a novel approach on how to investigate aggressive behavior in patients with psychotic disorders. Replication of our results in a bigger sample is needed to reliably develop a day-to-day risk assessment tool for aggressive behavior.

Applying a Dynamical Systems Model and Network Theory to Major Depressive Disorder.

Mental disorders like major depressive disorder can be modeled as complex dynamical systems. In this study we investigate the dynamic behavior of individuals to see whether or not we can expect a transition to another mood state. We introduce a mean field model to a binomial process, where we reduce a dynamic multidimensional system (stochastic cellular automaton) to a one-dimensional system to analyse the dynamics. Using maximum likelihood estimation, we can estimate the parameter of interest which, in combination with a bifurcation diagram, reflects the expectancy that someone has to transition to another mood state. After numerically illustrating the proposed method with simulated data, we apply this method to two empirical examples, where we show its use in a clinical sample consisting of patients diagnosed with major depressive disorder, and a general population sample. Results showed that the majority of the clinical sample was categorized as having an expectancy for a transition, while the majority of the general population sample did not have this expectancy. We conclude that the mean field model has great potential in assessing the expectancy for a transition between mood states. With some extensions it could, in the future, aid clinical therapists in the treatment of depressed patients.