Unsupervised Classification During Time-Series Model Building

Heterogeneity in suicide risk: Evidence from personalized dynamic models

Most theories of suicide propose within-person changes in psychological states cause suicidal thoughts/behaviors; however, most studies use between-person analyses. Thus, there are little empirical data exploring current theories in the way they are hypothesized to occur. We used a form of statistical modeling called group iterative multiple model estimation (GIMME) to explore one theory of suicide: The Interpersonal Theory of Suicide (IPTS). GIMME estimates personalized statistical models for each individual and associations shared across individuals. Data were from a real-time monitoring study of individuals with a history of suicidal thoughts/behavior (adult sample: participants = 111, observations = 25,242; adolescent sample: participants = 145, observations = 26,182). Across both samples, none of theorized IPTS effects (i.e., contemporaneous effect from hopeless to suicidal thinking) were shared at the group level. There was significant heterogeneity in the personalized models, suggesting there are different pathways through which different people come to experience suicidal thoughts/behaviors. These findings highlight the complexity of suicide risk and the need for more personalized approaches to assessment and prediction.

Dynamic processes in behavioral activation therapy for anhedonic adolescents: Modeling common and patient-specific relations

OBJECTIVE

Behavioral activation (BA) is a brief intervention for depression encouraging gradual and systematic re-engagement with rewarding activities and behaviors. Given this treatment focus, BA may be particularly beneficial for adolescents with prominent anhedonia, a predictor of poor treatment response and common residual symptom. We applied group iterative multiple model estimation (GIMME) to ecological momentary assessment (EMA) treatment data to investigate common and person-specific processes during BA for anhedonic adolescents.

METHOD

Thirty-nine adolescents (Mage = 15.7 years old, 67% female, 81% White) with elevated anhedonia (Snaith-Hamilton Pleasure Scale) were enrolled in a 12-week BA trial, with weekly anhedonia assessments. EMA surveys were triggered every other week (2-3 surveys per day) throughout treatment assessing current positive affect (PA) and negative affect (NA), engagement in pleasurable activities and social interactions, anticipatory pleasure, rumination, and recent pleasurable and stressful experiences.

RESULTS

A multilevel model revealed significant decreases in anhedonia, t(25.5) = -4.76, p < .001, over the 12-week trial. GIMME results indicated substantial heterogeneity in variable networks across patients. PA was the variable with the greatest number (22% of all paths vs. 11% for NA) of predictive paths to other symptoms (i.e., highest out-degree). Higher PA (but not NA) out-degree was associated with greater anhedonia improvement, t(25.8) = -2.22, p = .035.

CONCLUSIONS

Results revealed substantial heterogeneity in variable relations across patients, which may obscure the search for common processes of change in BA. PA may be a particularly important treatment target for anhedonic adolescents in BA. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Novel Approaches Toward Studying Change: Implications for Understanding and Treating Psychopathology

In this article, the authors critically evaluate contemporary models of psychopathology and therapies, underscoring the limitations of traditional symptom-based classification approaches in mental health. The authors introduce a paradigm shift in the field, toward a process-oriented and dynamic systems approach to psychotherapy that offers deeper insights into the complex interplay of symptoms and individual experiences in psychopathology. These approaches offer a more personalized and effective understanding and treatment of mental health issues, moving beyond static and 1-dimensional views. The authors discuss the implications for clinical practice, emphasizing improved assessment, diagnosis, and tailored treatment strategies.

Data-driven connectivity profiles relate to smoking cessation outcomes

At a group level, nicotine dependence is linked to differences in resting-state functional connectivity (rs-FC) within and between three large-scale brain networks: the salience network (SN), default mode network (DMN), and frontoparietal network (FPN). Yet, individuals may display distinct patterns of rs-FC that impact treatment outcomes. This study used a data-driven approach, Group Iterative Multiple Model Estimation (GIMME), to characterize shared and person-specific rs-FC features linked with clinically-relevant treatment outcomes. 49 nicotine-dependent adults completed a resting-state fMRI scan prior to a two-week smoking cessation attempt. We used GIMME to identify group, subgroup, and individual-level networks of SN, DMN, and FPN connectivity. Regression models assessed whether within- and between-network connectivity of individual rs-FC models was associated with baseline cue-induced craving, and craving and use of regular cigarettes (i.e., "slips") during cessation. As a group, participants displayed shared patterns of connectivity within all three networks, and connectivity between the SN-FPN and DMN-SN. However, there was substantial heterogeneity across individuals. Individuals with greater within-network SN connectivity experienced more slips during treatment, while individuals with greater DMN-FPN connectivity experienced fewer slips. Individuals with more anticorrelated DMN-SN connectivity reported lower craving during treatment, while SN-FPN connectivity was linked to higher craving. In conclusion, in nicotine-dependent adults, GIMME identified substantial heterogeneity within and between the large-scale brain networks. Individuals with greater SN connectivity may be at increased risk for relapse during treatment, while a greater positive DMN-FPN and negative DMN-SN connectivity may be protective for individuals during smoking cessation treatment.

Subgrouping with Chain Graphical VAR Models

Recent years have seen the emergence of an "idio-thetic" class of methods to bridge the gap between nomothetic and idiographic inference. These methods describe nomothetic trends in idiographic processes by pooling intraindividual information across individuals to inform group-level inference or vice versa. The current work introduces a novel "idio-thetic" model: the subgrouped chain graphical vector autoregression (scGVAR). The scGVAR is unique in its ability to identify subgroups of individuals who share common dynamic network structures in both lag(1) and contemporaneous effects. Results from Monte Carlo simulations indicate that the scGVAR shows promise over similar approaches when clusters of individuals differ in their contemporaneous dynamics and in showing increased sensitivity in detecting nuanced group differences while keeping Type-I error rates low. In contrast, a competing approach-the Alternating Least Squares VAR (ALS VAR) performs well when groups were separated by larger distances. Further considerations are provided regarding applications of the ALS VAR and scGVAR on real data and the strengths and limitations of both methods.

Like No Other? A Family-Specific Network Approach to Parenting Adolescents

Numerous theories suggest that parents and adolescents influence each other in diverse ways; however, whether these influences differ between subgroups or are unique to each family remains uncertain. Therefore, this study explored whether data-driven subgroups of families emerged that exhibited a similar daily interplay between parenting and adolescent affective well-being. To do so, Subgrouping Group Iterative Multiple Model Estimation (S-GIMME) was used to estimate family-specific dynamic network models, containing same- and next-day associations among five parenting practices (i.e., warmth, autonomy support, psychological control, strictness, monitoring) and adolescent positive and negative affect. These family-specific networks were estimated for 129 adolescents (Mage = 13.3, SDage = 1.2, 64% female, 87% Dutch), who reported each day on parenting and their affect for 100 consecutive days. The findings of S-GIMME did not identify data-driven subgroups sharing similar parenting-affect associations. Instead, each family displayed a unique pattern of temporal associations between the different practices and adolescent affect. Thus, the ways in which parenting practices were related to adolescents' affect in everyday life were family specific.

Clustering Analysis of Time Series of Affect in Dyadic Interactions

An important goal when analyzing multivariate time series is the identification of heterogeneity, both within and across individuals over time. This heterogeneity can represent different ways in which psychological processes manifest, either between people or within a person across time. In many instances, those differences can have systematic patterns that can be related to future outcomes. In close relationships, for example, the daily exchange of affect between two individuals in a couple can contain a particular structure that is different across people and can result in varying levels of relationship satisfaction. In this paper we use Louvain, a clustering method, as a tool to characterize heterogeneity in multivariate time series data. Using affect measures from dyadic interactions, we first determine that Louvain is adept at detecting homogeneous patterns that are distinct from one another. Additionally, these homogeneous points are linked, at some level, by time. Thus, we find that clustering via Louvain is useful to find time periods of stable, reoccurring patterns. However, using measures founded on information theory reveals that there is some level of information loss that is inevitable when clustering on levels of variable expression. Finally, we evaluate the predictive validity of the clustering method by examining the relation between the identified clusters of affect and measures outside the time series (i.e., relationship satisfaction and breakup taken one and two years later).

When Average Isn't Good Enough: Identifying Meaningful Subgroups in Clinical Data

Background

Clinical data are usually analyzed with the assumption that knowledge gathered from group averages applies to the individual. Doing so potentially obscures patients with meaningfully different trajectories of therapeutic change. Needed are "idionomic" methods that first examine idiographic patterns before nomothetic generalizations are made. The objective of this paper is to test whether such an idionomic method leads to different clinical conclusions.

Methods

51 patients completed weekly process measures and symptom severity over a period of eight weeks. Change trajectories were analyzed using a nomothetic approach and an idiographic approach with bottom-up clustering of similar individuals. The outcome was patients' well-being at post-treatment.

Results

Individuals differed in the extent that underlying processes were linked to symptoms. Average trend lines did not represent the intraindividual changes well. The idionomic approach readily identified subgroups of patients that differentially predicted distal outcomes (well-being).

Conclusions

Relying exclusively on average results may lead to an oversight of intraindividual pathways. Characterizing data first using idiographic approaches led to more refined conclusions, which is clinically useful, scientifically rigorous, and may help advance individualized psychotherapy approaches.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10608-023-10453-x.

Modeling Heterogeneity in Temporal Dynamics: Extending Latent State-Trait Autoregressive and Cross-lagged Panel Models to Mixture Distribution Models

Longitudinal models suited for the analysis of panel data, such as cross-lagged panel or autoregressive latent-state trait models, assume population homogeneity with respect to the temporal dynamics of the variables under investigation. This assumption is likely to be too restrictive in a myriad of research areas. We propose an extension of autoregressive and cross-lagged latent state-trait models to mixture distribution models. The models allow researchers to model unobserved person heterogeneity and qualitative differences in longitudinal dynamics based on comparatively few observations per person, while taking into account temporal dependencies between observations as well as measurement error in the variables. The models are extended to include categorical covariates, to investigate the distribution of encountered latent classes across observed groups. The potential of the models is illustrated with an application to self-esteem and affect data in patients with borderline personality disorder, an anxiety disorder, and healthy control participants. Requirements for the models' applicability are investigated in an extensive simulation study and recommendations for model applications are derived.

From the Individual to the Group: Using Idiographic Analyses and Two-Stage Random Effects Meta-Analysis to Obtain Population Level Inferences for within-Person Processes

In psychology, the use of portable technology and wearable devices to ease participant burden in data collection is on the rise. This creates increased interest in collecting real-time or near real-time data from individuals within their natural environments. As a result, vast amounts of observational time series data are generated. Often, motivation for collecting this data hinges on understanding within-person processes that underlie psychological phenomena. Motivated by the body of Dr. Peter Molenaar's life work calling for analytical approaches that consider potential heterogeneity and non-ergodicity, the focus of this paper is on using idiographic analyses to generate population inferences for within-person processes. Meta-analysis techniques using one-stage and two-stage random effects meta-analysis as implemented in single-case experimental designs are presented. The case for preferring a two-stage approach for meta-analysis of single-subject observational time series data is made and demonstrated using an empirical example. This provides a novel implementation of the methodology as prior implementations focus on applications to short time series with experimental designs. Inspired by Dr. Molenaar's work, we describe how an approach, two-stage random effects meta-analysis (2SRE-MA), aligns with recent calls to consider idiographic approaches when making population-level inferences regarding within-person processes.

Evaluating Discrete Time Methods for Subgrouping Continuous Processes

Rapid developments over the last several decades have brought increased focus and attention to the role of time scales and heterogeneity in the modeling of human processes. To address these emerging questions, subgrouping methods developed in the discrete-time framework-such as the vector autoregression (VAR)-have undergone widespread development to identify shared nomothetic trends from idiographic modeling results. Given the dependence of VAR-based parameters on the measurement intervals of the data, we sought to clarify the strengths and limitations of these methods in recovering subgroup dynamics under different measurement intervals. Building on the work of Molenaar and collaborators for subgrouping individual time-series by means of the subgrouped chain graphical VAR (scgVAR) and the subgrouping option in the group iterative multiple model estimation (S-GIMME), we present results from a Monte Carlo study aimed at addressing the implications of identifying subgroups using these discrete-time methods when applied to continuous-time data. Results indicate that discrete-time subgrouping methods perform well at recovering true subgroups when the measurement intervals are large enough to capture the full range of a system's dynamics, either via lagged or contemporaneous effects. Further implications and limitations are discussed therein.

Characterizing heterogeneity in early adolescent reward networks and individualized associations with behavioral and clinical outcomes

Associations between connectivity networks and behavioral outcomes such as depression are typically examined by comparing average networks between known groups. However, neural heterogeneity within groups may limit the ability to make inferences at the individual level as qualitatively distinct processes across individuals may be obscured in group averages. This study characterizes the heterogeneity of effective connectivity reward networks among 103 early adolescents and examines associations between individualized features and multiple behavioral and clinical outcomes. To characterize network heterogeneity, we used extended unified structural equation modeling to identify effective connectivity networks for each individual and an aggregate network. We found that an aggregate reward network was a poor representation of individuals, with most individual-level networks sharing less than 50% of the group-level network paths. We then used Group Iterative Multiple Model Estimation to identify a group-level network, subgroups of individuals with similar networks, and individual-level networks. We identified three subgroups that appear to reflect differences in network maturity, but this solution had modest validity. Finally, we found numerous associations between individual-specific connectivity features and behavioral reward functioning and risk for substance use disorders. We suggest that accounting for heterogeneity is necessary to use connectivity networks for inferences precise to the individual.

Blind Subgrouping of Task-based fMRI

Significant heterogeneity in network structures reflecting individuals' dynamic processes can exist within subgroups of people (e.g., diagnostic category, gender). This makes it difficult to make inferences regarding these predefined subgroups. For this reason, researchers sometimes wish to identify subsets of individuals who have similarities in their dynamic processes regardless of any predefined category. This requires unsupervised classification of individuals based on similarities in their dynamic processes, or equivalently, in this case, similarities in their network structures of edges. The present paper tests a recently developed algorithm, S-GIMME, that takes into account heterogeneity across individuals with the aim of providing subgroup membership and precise information about the specific network structures that differentiate subgroups. The algorithm has previously provided robust and accurate classification when evaluated with large-scale simulation studies but has not yet been validated on empirical data. Here, we investigate S-GIMME's ability to differentiate, in a purely data-driven manner, between brain states explicitly induced through different tasks in a new fMRI dataset. The results provide new evidence that the algorithm was able to resolve, in an unsupervised data-driven manner, the differences between different active brain states in empirical fMRI data to segregate individuals and arrive at subgroup-specific network structures of edges. The ability to arrive at subgroups that correspond to empirically designed fMRI task conditions, with no biasing or priors, suggests this data-driven approach can be a powerful addition to existing methods for unsupervised classification of individuals based on their dynamic processes.

Adolescent functional network connectivity prospectively predicts adult anxiety symptoms related to perceived COVID-19 economic adversity

BACKGROUND

Stressful events, such as the COVID-19 pandemic, are major contributors to anxiety and depression, but only a subset of individuals develop psychopathology. In a population-based sample (N = 174) with a high representation of marginalized individuals, this study examined adolescent functional network connectivity as a marker of susceptibility to anxiety and depression in the context of adverse experiences.

METHODS

Data-driven network-based subgroups were identified using an unsupervised community detection algorithm within functional neural connectivity. Neuroimaging data collected during emotion processing (age 15) were extracted from a priori regions of interest linked to anxiety and depression. Symptoms were self-reported at ages 15, 17, and 21 (during COVID-19). During COVID-19, participants reported on pandemic-related economic adversity. Differences across subgroup networks were first examined, then subgroup membership and subgroup-adversity interaction were tested to predict change in symptoms over time.

RESULTS

Two subgroups were identified: Subgroup A, characterized by relatively greater neural network variation (i.e., heterogeneity) and density with more connections involving the amygdala, subgenual cingulate, and ventral striatum; and the more homogenous Subgroup B, with more connections involving the insula and dorsal anterior cingulate. Accounting for initial symptoms, subgroup A individuals had greater increases in symptoms across time (β = .138, p = .042), and this result remained after adjusting for additional covariates (β = .194, p = .023). Furthermore, there was a subgroup-adversity interaction: compared with Subgroup B, Subgroup A reported greater anxiety during the pandemic in response to reported economic adversity (β = .307, p = .006), and this remained after accounting for initial symptoms and many covariates (β = .237, p = .021).

CONCLUSIONS

A subgrouping algorithm identified young adults who were susceptible to adversity using their personalized functional network profiles derived from a priori brain regions. These results highlight potential prospective neural signatures involving heterogeneous emotion networks that predict individuals at the greatest risk for anxiety when experiencing adverse events.

Resting-state connectivity underlying cognitive control's association with perspective taking in callous-unemotional traits

Callous-Unemotional (CU) traits are often associated with impairments in perspective taking and cognitive control (regulating goal directed behavior); and adolescents with CU traits demonstrate aberrant brain activation/connectivity in areas underlying these processes. Together cognitive control and perspective taking are thought to link mechanistically to explain CU traits. Because increased cognitive control demands modulate perspective taking ability among both typically developing samples and individuals with elevated CU traits, understanding the neurophysiological substrates of these constructs could inform efforts to alleviate societal costs of antisocial behavior. The present study uses GIMME to examine the heterogenous functional brain properties (i.e., connection density, node centrality) underlying cognitive control's influence on perspective taking among adolescents on a CU trait continuum. Results reveal that cognitive control had a negative indirect association with CU traits via perspective taking; and brain connectivity indirectly associated with lower CU traits - specifically the social network via perspective taking and conflict network via cognitive control. Additionally, less negative connection density between the social and conflict networks was directly associated with higher CU traits. Our results support the growing literature on cognitive control's influence on socio-cognitive functioning in CU traits and extends that work by identifying underlying functional brain properties.

The influence of autoregressive relation strength and search strategy on directionality recovery in group iterative multiple model estimation

Unified structural equation modeling (uSEM) implemented in the group iterative multiple model estimation (GIMME) framework has recently been widely used for characterizing within-person network dynamics of behavioral and functional neuroimaging variables. Previous studies have established that GIMME accurately recovers the presence of relations between variables. However, recovery of relation directionality is less consistent, which is concerning given the importance of directionality estimates for many research questions. There is evidence that strong autoregressive relations may aid directionality recovery and indirect evidence that a novel version of GIMME allowing for multiple solutions could improve recovery when such relations are weak, but it remains unclear how these strategies perform under a range of study conditions. Using comprehensive simulations that varied the strength of autoregressive relations among other factors, this study evaluated the directionality recovery of two GIMME search strategies: (a) estimating autoregressive relations by default in the null model (GIMME-AR) and (b) generating multiple solution paths (GIMME-MS). Both strategies recovered directionality best-and were roughly equivalent in performance-when autoregressive relations were strong (e.g., β = .60). When they were weak (β ≤ .10), GIMME-MS displayed an advantage, although overall directionality recovery was modest. Analyses of empirical data in which autoregressive relations were characteristically strong (resting state functional MRI) versus weak (daily diary) mirrored simulation results and confirmed that these strategies can disagree on directionality when autoregressive relations are weak. Findings have important implications for psychological and neuroimaging applications of uSEM/GIMME and suggest specific scenarios in which researchers might or might not be confident in directionality results. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Social goals in girls transitioning to adolescence: associations with psychopathology and brain network connectivity

The motivation to socially connect with peers increases during adolescence in parallel with changes in neurodevelopment. These changes in social motivation create opportunities for experiences that can impact risk for psychopathology, but the specific motivational presentations that confer greater psychopathology risk are not fully understood. To address this issue, we used a latent profile analysis to identify the multidimensional presentations of self-reported social goals in a sample of 220 girls (9-15 years old, M = 11.81, SD = 1.81) that was enriched for internalizing symptoms, and tested the association between social goal profiles and psychopathology. Associations between social goals and brain network connectivity were also examined in a subsample of 138 youth. Preregistered analyses revealed four unique profiles of social goal presentations in these girls. Greater psychopathology was associated with heightened social goals such that higher clinical symptoms were related to a greater desire to attain social competence, avoid negative feedback and gain positive feedback from peers. The profiles endorsing these excessive social goals were characterized by denser connections among social-affective and cognitive control brain regions. These findings thus provide preliminary support for adolescent-onset changes in motivating factors supporting social engagement that may contribute to risk for psychopathology in vulnerable girls.

Resting-state connectivity subtype of comorbid PTSD and alcohol use disorder moderates improvement from integrated prolonged exposure therapy in Veterans

BACKGROUND

Posttraumatic stress disorder (PTSD) and alcohol use disorder (AUD) are highly comorbid and are associated with significant functional impairment and inconsistent treatment outcomes. Data-driven subtyping of this clinically heterogeneous patient population and the associated underlying neural mechanisms are highly needed to identify who will benefit from psychotherapy.

METHODS

In 53 comorbid PTSD/AUD patients, resting-state functional magnetic resonance imaging was collected prior to undergoing individual psychotherapy. We used a data-driven approach to subgroup patients based on directed connectivity profiles. Connectivity subgroups were compared on clinical measures of PTSD severity and heavy alcohol use collected at pre- and post-treatment.

RESULTS

We identified a subgroup of patients associated with improvement in PTSD symptoms from integrated-prolonged exposure therapy. This subgroup was characterized by lower insula to inferior parietal cortex (IPC) connectivity, higher pregenual anterior cingulate cortex (pgACC) to posterior midcingulate cortex connectivity and a unique pgACC to IPC path. We did not observe any connectivity subgroup that uniquely benefited from integrated-coping skills or subgroups associated with change in alcohol consumption.

CONCLUSIONS

Data-driven approaches to characterize PTSD/AUD subtypes have the potential to identify brain network profiles that are implicated in the benefit from psychological interventions - setting the stage for future research that targets these brain circuit communication patterns to boost treatment efficacy.

Distal neuropathic pain in HIV is associated with functional connectivity patterns in default mode and salience networks

HIV-associated distal neuropathic pain (DNP) is one of the most prevalent, disabling, and treatment-resistant complications of HIV, but its biological underpinnings are incompletely understood. While data specific to mechanisms underlying HIV DNP are scarce, functional neuroimaging of chronic pain more broadly implicates the role of altered resting-state functional connectivity within and between salience network (SN) and default mode network (DMN) regions. However, it remains unclear the extent to which HIV DNP is associated with similar alterations in connectivity. The current study aimed to bridge this gap in the literature through examination of resting-state functional connectivity patterns within SN and DMN regions among people with HIV (PWH) with and without DNP. Resting state functional magnetic resonance imaging (rs-fMRI) scans were completed among 62 PWH with HIV-associated peripheral neuropathy, of whom 27 reported current DNP and 35 did not. Using subgrouping group iterative multiple estimation, we compared connectivity patterns in those with current DNP to those without. We observed weaker connectivity between the medial prefrontal cortex (MPFC) and posterior cingulate cortex (PCC) and stronger connectivity between the anterior cingulate cortex (ACC) and thalamus among those reporting DNP. Overall, these findings implicate altered within DMN (i.e., MPFC-PCC) and within SN (i.e., ACC-thalamus) connectivity as potential manifestations of adaptation to pain from neuropathy and/or mechanisms underlying the development/maintenance of DNP. Findings are discussed in the context of differential brain response to pain (i.e., mind wandering, pain aversion, pain facilitation/inhibition) and therapeutic implications.

Network Physiology of Exercise: Beyond Molecular and Omics Perspectives

Molecular Exercise Physiology and Omics approaches represent an important step toward synthesis and integration, the original essence of Physiology. Despite the significant progress they have introduced in Exercise Physiology (EP), some of their theoretical and methodological assumptions are still limiting the understanding of the complexity of sport-related phenomena. Based on general principles of biological evolution and supported by complex network science, this paper aims to contrast theoretical and methodological aspects of molecular and network-based approaches to EP. After explaining the main EP challenges and why sport-related phenomena cannot be understood if reduced to the molecular level, the paper proposes some methodological research advances related to the type of studied variables and measures, the data acquisition techniques, the type of data analysis and the assumed relations among physiological levels. Inspired by Network Physiology, Network Physiology of Exercise provides a new paradigm and formalism to quantify cross-communication among diverse systems across levels and time scales to improve our understanding of exercise-related phenomena and opens new horizons for exercise testing in health and disease.

Daily Contributors of Tinnitus Loudness and Distress: An Ecological Momentary Assessment Study

Background

Tinnitus is a heterogeneous condition which may be associated with moderate to severe disability, but the reasons why only a subset of individuals is burdened by the condition are not fully clear. Ecological momentary assessment (EMA) allows a better understanding of tinnitus by capturing the fluctuations of tinnitus symptoms, such as distress and loudness, and psychological processes, such as emotional arousal, overall stress, mood, and concentration and how these variables interact over time. Whether any of those variables have an influence over the next day, that is, whether any of these variables are auto- or cross-correlated, is still unanswered.

Objectives

Assess whether behavioral and symptom-related data from tinnitus users from the TrackYourTinnitus (TYT) mobile app have an impact on tinnitus loudness and distress on subsequent days.

Methods

Anonymized data was collected from 278 users of the iOS or Android TYT apps between 2014 and 2020. Tinnitus-related distress, tinnitus loudness, concentration level, mood, emotional arousal, and overall stress level were assessed using either a slider or the Wong-Baker Pain FACES scale via a daily survey. Three modeling strategies were used to investigate whether tinnitus loudness and distress are affected by previous days symptoms or psychological processes: auto- and cross correlations, regressions with elastic net regularization, and subgrouping within group iterative multiple model estimation (S-GIMME).

Results

No autocorrelation or cross-correlation was observed at the group level between the variables assessed. However, application of the regression models with elastic net regularization identified individualized predictors of tinnitus loudness and distress for most participants, with the models including contemporaneous and lagged information from the previous day. S-GIMME corroborated these findings by identifying individualized predictors of tinnitus loudness and distress from the previous day.

Discussion

We showed that tinnitus loudness and tinnitus distress are affected by the contemporaneous and lagged dynamics of behavioral and emotional processes measured through EMA. These effects were seen at the group, and individual levels. The relevance EMA and the implications of the insights derived from it for tinnitus care are discussed, especially considering current trends toward the individualization of tinnitus care.

Heterogeneity in PFC-amygdala connectivity in middle childhood, and concurrent interrelations with inhibitory control and anxiety symptoms

The prefrontal cortex (PFC) is a key brain area in considering adaptive regulatory behaviors. This includes regulatory projections to regions of the limbic system such as the amygdala, where the nature of functional connections may confer lower risk for anxiety disorders. The PFC is also associated with behaviors like executive functioning. Inhibitory control is a behavior encompassed by executive functioning and is generally viewed favorably for adaptive socioemotional development. Yet, some research suggests that high levels of inhibitory control may actually be a risk factor for some maladaptive developmental outcomes, like anxiety disorders. In a sample of 51 children ranging from 7 to 9 years old, we examined resting state functional connectivity between regions of the PFC and the amygdala. We used Subgrouping Group Iterative Multiple Model Estimation (S-GIMME) to identify and characterize data-driven subgroups of individuals with similar networks of connectivity between these brain regions. Generated subgroups were collapsed into children characterized by the presence or absence of recovered connections between the PFC and amygdala. For subsets of children with available data (N = 38-44), we then tested whether inhibitory control, as measured by a stop signal task, moderated the relation between these subgroups and child-reported anxiety symptoms. We found an inverse relation between stop-signal reaction times and reported count of anxiety symptoms when covarying for connectivity group, suggesting that greater inhibitory control was actually related to greater anxiety symptoms, but only when accounting for patterns of PFC-amygdala connectivity. These data suggest that there is a great deal of heterogeneity in the nature of functional connections between the PFC and amygdala during this stage of development. The findings also provide support for the notion of high levels of inhibitory control as a risk factor for anxiety, but trait-level biopsychosocial factors may be important to consider in assessing the nature of risk.

Understanding Pain and Trauma Symptoms in Veterans From Resting-State Connectivity: Unsupervised Modeling

Trauma and posttraumatic stress are highly comorbid with chronic pain and are often antecedents to developing chronic pain conditions. Pain and trauma are associated with greater utilization of medical services, greater use of psychiatric medication, and increased total cost of treatment. Despite the high overlap in the clinic, the neural mechanisms of pain and trauma are often studied separately. In this study, resting-state functional magnetic resonance imaging (rs-fMRI) scans were completed among a diagnostically heterogeneous sample of veterans with a range of back pain and trauma symptoms. Using Group Iterative Multiple Model Estimation (GIMME), an effective functional connectivity analysis, we explored an unsupervised model deriving subgroups based on path similarity in a priori defined regions of interest (ROIs) from brain regions implicated in the experience of pain and trauma. Three subgroups were identified by patterns in functional connection and differed significantly on several psychological measures despite similar demographic and diagnostic characteristics. The first subgroup was highly connected overall, was characterized by functional connectivity from the nucleus accumbens (NAc), the anterior cingulate cortex (ACC), and the posterior cingulate cortex (PCC) to the insula and scored low on pain and trauma symptoms. The second subgroup did not significantly differ from the first subgroup on pain and trauma measures but was characterized by functional connectivity from the ACC and NAc to the thalamus and from ACC to PCC. The third subgroup was characterized by functional connectivity from the thalamus and PCC to NAc and scored high on pain and trauma symptoms. Our results suggest that, despite demographic and diagnostic similarities, there may be neurobiologically dissociable biotypes with different mechanisms for managing pain and trauma. These findings may have implications for the determination of appropriate biotype-specific interventions that target these neurological systems.

Unsupervised classification reveals consistency and degeneracy in neural network patterns of emotion

In the present study, we used an unsupervised classification algorithm to reveal both consistency and degeneracy in neural network connectivity during anger and anxiety. Degeneracy refers to the ability of different biological pathways to produce the same outcomes. Previous research is suggestive of degeneracy in emotion, but little research has explicitly examined whether degenerate functional connectivity patterns exist for emotion categories such as anger and anxiety. Twenty-four subjects underwent functional magnetic resonance imaging (fMRI) while listening to unpleasant music and self-generating experiences of anger and anxiety. A data-driven model building algorithm with unsupervised classification (subgrouping Group Iterative Multiple Model Estimation) identified patterns of connectivity among 11 intrinsic networks that were associated with anger vs anxiety. As predicted, degenerate functional connectivity patterns existed within these overarching consistent patterns. Degenerate patterns were not attributable to differences in emotional experience or other individual-level factors. These findings are consistent with the constructionist account that emotions emerge from flexible functional neuronal assemblies and that emotion categories such as anger and anxiety each describe populations of highly variable instances.

Person-specific connectivity mapping uncovers differences of bilingual language experience on brain bases of attention in children

Bilingualism influences children's cognition, yet bilinguals vary greatly in their dual-language experiences. To uncover sources of variation in bilingual and monolingual brain function, the present study used standard analysis and innovative person-specific connectivity models combined with a data-driven grouping algorithm. Children (ages 7-9; N = 52) completed a visuo-spatial attention task while undergoing functional near-infrared spectroscopy neuroimaging. Both bilingual and monolingual groups performed similarly, and engaged bilateral frontal and parietal regions. However, bilinguals showed greater brain activity than monolinguals in left frontal and parietal regions. Connectivity models revealed two empirically-derived subgroups. One subgroup was composed of monolinguals and bilinguals who were more English dominant, and showed left frontal-parietal connections. The other was composed of bilinguals who were balanced in their dual-language abilities and showed left frontal lobe connections. The findings inform how individual variation in early language experiences influences children's emerging cortical networks for executive function, and reveal efficacy of data-driven approaches.

Neural connectivity underlying adolescent social learning in sibling dyads

Social learning theory posits that adolescents learn to adopt social norms by observing the behaviors of others and internalizing the associated outcomes. However, the underlying neural processes by which social learning occurs is less well-understood, despite extensive neurobiological reorganization and a peak in social influence sensitivity during adolescence. Forty-four adolescents (Mage = 12.2 years) completed an fMRI scan while observing their older sibling within four years of age (Mage = 14.3 years) of age complete a risky decision-making task. Group iterative multiple model estimation (GIMME) was used to examine patterns of directional brain region connectivity supporting social learning. We identified group-level neural pathways underlying social observation including the anterior insula to the anterior cingulate cortex and mentalizing regions to social cognition regions. We also found neural states based on adolescent sensitivity to social learning via age, gender, modeling, differentiation, and behavior. Adolescents who were more likely to be influenced elicited neurological up-regulation whereas adolescents who were less likely to be socially influenced elicited neurological down-regulation during risk-taking. These findings highlight patterns of how adolescents process information while a salient influencer takes risks, as well as salient neural pathways that are dependent on similarity factors associated with social learning theory.

Neural heterogeneity underlying late adolescent motivational processing is linked to individual differences in behavioral sensation seeking

Adolescent risk-taking, including sensation seeking (SS), is often attributed to developmental changes in connectivity among brain regions implicated in cognitive control and reward processing. Despite considerable scientific and popular interest in this neurodevelopmental framework, there are few empirical investigations of adolescent functional connectivity, let alone examinations of its links to SS behavior. The studies that have been done focus on mean-based approaches and leave unanswered questions about individual differences in neurodevelopment and behavior. The goal of this paper is to take a person-specific approach to the study of adolescent functional connectivity during a continuous motivational state, and to examine links between connectivity and self-reported SS behavior in 104 adolescents (MAge  = 19.3; SDAge  = 1.3). Using Group Iterative Multiple Model Estimation (GIMME), person-specific connectivity during two neuroimaging runs of a monetary incentive delay task was estimated among 12 a priori brain regions of interest representing reward, cognitive, and salience networks. Two data-driven subgroups were detected, a finding that was consistent between both neuroimaging runs, but associations with SS were only found in the first run, potentially reflecting neural habituation in the second run. Specifically, the subgroup that had unique connections between reward-related regions had greater SS and showed a distinctive relation between connectivity strength in the reward regions and SS. These findings provide novel evidence for heterogeneity in adolescent brain-behavior relations by showing that subsets of adolescents have unique associations between neural motivational processing and SS. Findings have broader implications for future work on reward processing, as they demonstrate that brain-behavior relations may attenuate across runs.

Averting the Next Credibility Crisis in Psychological Science: Within-Person Methods for Personalized Diagnostics and Intervention

Personalizing assessments, predictions, and treatments of individuals is currently a defining trend in psychological research and applied fields, including personalized learning, personalized medicine, and personalized advertisement. For instance, the recent pandemic has reminded parents and educators of how challenging yet crucial it is to get the right learning task to the right student at the right time. Increasingly, psychologists and social scientists are realizing that the between-person methods that we have long relied upon to describe, predict, and treat individuals may fail to live up to these tasks (e.g., Molenaar, 2004). Consequently, there is a risk of a credibility loss, possibly similar to the one seen during the replicability crisis (Ioannides, 2005), because we have only started to understand how many of the conclusions that we tend to draw based on between-person methods are based on a misunderstanding of what these methods can tell us and what they cannot. An imminent methodological revolution will likely lead to a change of even well-established psychological theories (Barbot et al., 2020). Fortunately, methodological solutions for personalized descriptions and predictions, such as many within-person analyses, are available and undergo rapid development, although they are not yet embraced in all areas of psychology, and some come with their own limitations. This article first discusses the extent of the theory-method gap, consisting of theories about within-person patterns being studied with between-person methods in psychology, and the potential loss of trust that might follow from this theory-method gap. Second, this article addresses advantages and limitations of available within-person methods. Third, this article discusses how within-person methods may help improving the individual descriptions and predictions that are needed in many applied fields that aim for tailored individual solutions, including personalized learning and personalized medicine.

Early Life Stress and Neurodevelopment in Adolescence: Implications for Risk and Adaptation

An alarming high proportion of youth experience at least one kind of stressor in childhood and/or adolescence. Exposure to early life stress is associated with increased risk for psychopathology, accelerated biological aging, and poor physical health; however, it is important to recognize that not all youth who experience such stress go on to develop difficulties. In fact, resilience, or positive adaptation in the face of adversity, is relatively common. Individual differences in vulnerability or resilience to the effects of early stress may be represented in the brain as specific patterns, profiles, or signatures of neural activation, structure, and connectivity (i.e., neurophenotypes). Whereas neurophenotypes of risk that reflect the deleterious effects of early stress on the developing brain are likely to exacerbate negative outcomes in youth, neurophenotypes of resilience may reduce the risk of experiencing these negative outcomes and instead promote positive functioning. In this chapter we describe our perspective concerning the neurobiological mechanisms and moderators of risk and resilience in adolescence following early life stress and integrate our own work into this framework. We present findings suggesting that exposure to stress in childhood and adolescence is associated with functional and structural alterations in neurobiological systems that are important for social-affective processing and for cognitive control. While some of these neurobiological alterations increase risk for psychopathology, they may also help to limit adolescents' sensitivity to subsequent negative experiences. We also discuss person-centered strategies that we believe can advance our understanding of risk and resilience to early stress in adolescents. Finally, we describe ways in which the field can broaden its focus to include a consideration of other types of environmental factors, such as environmental pollutants, in affecting both risk and resilience to stress-related health difficulties in youth.

Building on the past 50 years, not starting over: A balanced interpretation of meta-analyses, reviews, and commentaries on treatments for suicide and self-injury

Several recent meta-analyses on interventions for self-injurious thoughts and behaviors (SITBs) have been conducted. The primary finding of these meta-analyses is that the observed effects of interventions for SITBs are generally quite small and are far from where we need to be as a field. Although we agree with these general findings, we disagree, however, with many of the overly bleak conclusions drawn from these findings that emphasize creating new treatments while discounting the benefit of improving existing interventions and the decades of research that were involved in creating them. Accordingly, we offer three future directions with promise to build upon and improve our existing treatments, while we simultaneously work to develop new ones: (1) determine which intervention(s) are needed for which person and at which time, (2) conduct more research on intervention length before concluding that brief interventions are just as efficacious as longer ones, and (3) evaluate the potential of comprehensive models of suicide prevention as a more efficacious alternative to any one individual intervention.

Using temporal network methods to reveal the idiographic nature of development

Averages dominate developmental science: There are representative groups, mean trajectories, and generalizations to typical children. Nearly all parents and teachers, however, eagerly proclaim that few youth are average; each child, adolescent, and young adult is unique. Indeed, individual youth are the focus of many eminent developmental theories, yet there is a shocking paucity of developmental methods-including study designs and analysis techniques-that truly afford individual-level inferences. Thus, the goal of this chapter is to explicate the advantages of an idiographic approach to developmental science, that is, an approach that provides insight into individual youth, often by studying within-person variation in intensive longitudinal data, such as densely coded observations, repeated daily or momentary assessments, and functional neuroimages. In three domains across development, the chapter illustrates the benefits of an idiographic approach by comparing empirical conclusions offered by traditional mean-based analysis techniques versus techniques that leverage the temporal and individualized nature of intensive longitudinal data. The chapter then concentrates on group iterative multiple model estimation (GIMME), which is an analysis technique that uses intensive longitudinal data to create youth-specific temporal networks, detailing how brain regions or behaviors are directionally related across time. The promise of GIMME is exemplified by applications to three different domains across development. The chapter closes by encouraging future idiographic developmental science to consider how research questions, study designs, and data analyses can be formed, implemented, and conducted in ways that optimize inferences about individual-not average-youth.

Applying Deep Learning in the Training of Communication Design Talents Under University-Industrial Research Collaboration

The purpose of the study was to solve the problem of the mismatching between the supply and demand of the talents that universities provide for society, whose major is communication design. The correlations between social post demand and university cultivation, as well as between social post demand and the demand indexes of enterprises for posts, are explored under the guidance of University-Industrial Research Collaboration. The backpropagation neural network (BPNN) is used, and the advantages of the Seasonal Autoregressive Integrated Moving Average model (SARIMA) model are combined to design the SARIMA-BPNN (SARIMA-BP) model after the relevant parameters are adjusted. Through the experimental analysis, it is found that the error of the root mean square of the designed SARIMA-BP model in post prediction is 7.523 and that of the BPNN model is 16.122. The effect of the prediction model that was designed based on deep learning is smaller than that of the previous model based on the neural network, and it can predict future posts more accurately for colleges and universities. Guided by the “University-Industrial Research Collaboration,” students will have more practice in the teaching process in response to social needs. “University-Industrial Research Collaboration” guides the teaching direction for communication design majors and can help to cultivate communication design talents who are competent for the post provided.

Neural connectivity biotypes: associations with internalizing problems throughout adolescence

BACKGROUND

Neurophysiological patterns may distinguish which youth are at risk for the well-documented increase in internalizing symptoms during adolescence. Adolescents with internalizing problems exhibit altered resting-state functional connectivity (RSFC) of brain regions involved in socio-affective processing. Whether connectivity-based biotypes differentiate adolescents' levels of internalizing problems remains unknown.

METHOD

Sixty-eight adolescents (37 females) reported on their internalizing problems at ages 14, 16, and 18 years. A resting-state functional neuroimaging scan was collected at age 16. Time-series data of 15 internalizing-relevant brain regions were entered into the Subgroup-Group Iterative Multi-Model Estimation program to identify subgroups based on RSFC maps. Associations between internalizing problems and connectivity-based biotypes were tested with regression analyses.

RESULTS

Two connectivity-based biotypes were found: a Diffusely-connected biotype (N = 46), with long-range fronto-parietal paths, and a Hyper-connected biotype (N = 22), with paths between subcortical and medial frontal areas (e.g. affective and default-mode network regions). Higher levels of past (age 14) internalizing problems predicted a greater likelihood of belonging to the Hyper-connected biotype at age 16. The Hyper-connected biotype showed higher levels of concurrent problems (age 16) and future (age 18) internalizing problems.

CONCLUSIONS

Differential patterns of RSFC among socio-affective brain regions were predicted by earlier internalizing problems and predicted future internalizing problems in adolescence. Measuring connectivity-based biotypes in adolescence may offer insight into which youth face an elevated risk for internalizing disorders during this critical developmental period.

Resting-state network topology characterizing callous-unemotional traits in adolescence

BACKGROUND

Callous-unemotional (CU) traits, a youth antisocial phenotype, are hypothesized to associate with aberrant connectivity (dis-integration) across the salience (SAL), default mode (DMN), and frontoparietal (FPN) networks. However, CU traits have a heterogeneous presentation and previous research has not modeled individual heterogeneity in resting-state connectivity amongst adolescents with CU traits. The present study models individual-specific network maps and examines topological features of individual and subgroup maps in relation to CU traits.

METHODS

Participants aged 13-17 (n = 84, male = 55%, female = 45%) completed resting-state functional connectivity and the inventory of callous-unemotional traits as part of the Nathan Klein Rockland study. A sparse network approach (GIMME) was used to derive individual-level and subgroup maps of all participants. We then examined heterogeneous network features, including positive and negative connection density, associated with CU traits.

RESULTS

Higher rates of CU traits increased probability of inclusion in one subgroup, which had the highest mean level of CU traits. Analysis of network features reveals less density (positive and negative) within the FPN and greater density between DMN-FPN associated with CU traits.

DISCUSSION

Findings indicate heterogeneous person-specific connections and some subgroup connections amongst adolescents associate with CU traits. Higher CU traits associate with lower density in the FPN, which has been associated with attention and inhibition, and higher density between the DMN-FPN, which have been linked with cognitive control, social working memory, and empathy. Our findings suggest less efficiency in FPN function which, when considered mechanistically, could result in difficulty suppressing DMN when task positive networks are engaged. This is an area for further exploration but could explain cognitive and socio-affective impairments in CU traits.

Resting state functional connectivity subtypes predict discrete patterns of cognitive-affective functioning across levels of analysis among patients with treatment-resistant depression

Resting state functional connectivity (RSFC) in ventral affective (VAN), default mode (DMN) and cognitive control (CCN) networks may partially underlie heterogeneity in depression. The current study used data-driven parsing of RSFC to identify subgroups of patients with treatment-resistant depression (TRD; n = 70) and determine if subgroups generalized to transdiagnostic measures of cognitive-affective functioning relevant to depression (indexed across self-report, behavioral, and molecular levels of analysis). RSFC paths within key networks were characterized using Subgroup-Group Iterative Multiple Model Estimation. Three connectivity-based subgroups emerged: Subgroup A, the largest subset and containing the fewest pathways; Subgroup B, containing unique bidirectional VAN/DMN negative feedback; and Subgroup C, containing the most pathways. Compared to other subgroups, subgroup B was characterized by lower self-reported positive affect and subgroup C by higher self-reported positive affect, greater variability in induced positive affect, worse response inhibition, and reduced striatal tissue iron concentration. RSFC-based categorization revealed three TRD subtypes associated with discrete aberrations in transdiagnostic cognitive-affective functioning that were largely unified across levels of analysis and were maintained after accounting for the variability captured by a disorder-specific measure of depressive symptoms. Findings advance understanding of transdiagnostic brain-behavior heterogeneity in TRD and may inform novel treatment targets for this population.

Dynamic Mixture Modeling with dynr

Mixture modeling is commonly used to model sample heterogeneity by identifying unobserved classes of individuals with similar characteristics. Despite abundance of evidence in the literature suggesting that individuals are often characterized by different dynamic processes underlying their physiological, cognitive, psychological, and behavioral states, applications of dynamic mixture modeling are surprisingly lacking. We present here a proof-of-concept example of dynamic mixture modeling, where latent groups of individuals were identified based on different dynamic patterns in their time series. Our sample consists of 192 men who were in a heterosexual relationship. They were asked to complete a daily questionnaire involving emotions related to their relationship. Two latent groups were identified based on the strength of association between positive (e.g., loving) and negative (e.g., doubtful) affect. Men in the group characterized by a strong negative association (β=-.67) tended to be younger and had higher levels of anxiety toward their relationship than men in the other group, which was characterized by a weaker negative association (β=-.31). We illustrate the specification and estimation of dynamic mixture model using "dynr," an R package capable of handling a broad class of linear and nonlinear discrete- and continuous-time models with regime-switching properties.

Neural Changes in Reward Processing Following Approach Avoidance Training for Depression

Altered approach motivation is hypothesized to be critical for the maintenance of depression. Computer-administered approach-avoidance training programs to increase approach action tendencies toward positive stimuli produce beneficial outcomes. However, there have been few studies examining neural changes following approach-avoidance training. Participants with Major Depressive Disorder were randomized to an Approach Avoidance Training (AAT) manipulation intended to increase approach tendencies for positive social cues (n=13) or a control procedure (n=15). We examined changes in neural activation (primary outcome) and connectivity patterns using Group Iterative Multiple Model Estimation during a social reward anticipation task (exploratory). A laboratory-based social affiliation task was also administered following the manipulation to measure affect during anticipation of real-world social activity. Individuals in the AAT group demonstrated increased activation in reward processing regions during social reward anticipation relative to the control group from pre to post-training. Following training, connectivity patterns across reward regions were observed in the full sample and connectivity between the medial PFC and caudate was associated with anticipatory positive affect before the social interaction; preliminary evidence of differential connectivity patterns between the two groups also emerged. Results support models whereby modifying approach-oriented behavioral tendencies with computerized training leads to alterations in reward circuitry. (NCT02330744).

Individualized learning potential in stressful times: How to leverage intensive longitudinal data to inform online learning

Societal events - such as natural disasters, political shifts, or economic downturns - are time-varying and impact the learning potential of students in unique ways. These impacts are likely accentuated during the COVID-19 pandemic, which precipitated an abrupt and wholesale transition to online education. Unfortunately, the individual-level consequences of these events are difficult to determine because the extant literature focuses on single-occasion surveys that produce only group-level inferences. To better understand individual-level variability in stress and learning, intensive longitudinal data can be leveraged. The goal of this paper is to illustrate this by discussing three different techniques for the analysis of intensive longitudinal data: (1) regression analyses; (2) multilevel models; and (3) person-specific network models, (e.g., group iterative multiple model estimation; GIMME). For each technique, a brief background in the context of education research is provided, an illustrative analysis is presented using data from college students who completed a 75-day intensive longitudinal study of cognition, somatic symptoms, anxiety, and intellectual interests during the 2016 U.S. Presidential election - a period of heightened sociopolitical stress - and strengths and limitations are considered. The paper ends with recommendations for future research, especially for intensive longitudinal studies of online education during COVID-19.

Specifying exogeneity and bilinear effects in data-driven model searches

Data-driven model searches provide the opportunity to quantify person-specific processes using ambulatory assessment data. Here, the search space typically includes all potential relations among variables, meaning that all variables can potentially explain variability in all other variables. Oftentimes, this is unrealistic. For example, weather is unlikely to be predicted by someone's emotional state, whereas the reverse might be true. Allowing for specification of exogenous variables, or variables that are not predicted within the system, permits more realistic models and allows the researcher to model contextual change processes via the use of moderation variables. We use two sets of daily diary data to demonstrate the capabilities of allowing for the specification of exogenous variables in GIMME (Group Iterative Multiple Model Estimation), a model search algorithm that allows for models with idiographic, individual-level as well as subgroup- and group-level processes with intensive longitudinal data. First, using data collected from individuals diagnosed with personality disorders, we show results where weather-related and temporal basis variables are specified as exogenous, and reports on affect and behavior are endogenous. Next, we demonstrate the modeling of treatment effects in an intervention study, looking at data from a 6-week meditation workshop in midlife adults. Finally, we use the meditation intervention data to demonstrate modeling moderation effects, where relationships between two endogenous variables are dependent on the current stage of the study for a given participant (i.e., currently attending meditation classes or not). We end by presenting adaptive LASSO as a method for probing results obtained from GIMME.

Path and Directionality Discovery in Individual Dynamic Models: A Regularized Unified Structural Equation Modeling Approach for Hybrid Vector Autoregression

There recently has been growing interest in the study of psychological and neurological processes at an individual level. One goal in such endeavors is to construct person-specific dynamic assessments using time series techniques such as Vector Autoregressive (VAR) models. However, two problems exist with current VAR specifications: (1) VAR models are restricted in that contemporaneous relations are typically modeled either as undirected relations among residuals or directed relations among observed variables, but not both; (2) current estimation frameworks are limited by the reliance on stepwise model building procedures. This study adopts a new modeling approach. We first extended the current unified SEM (uSEM) framework, a widely used structural VAR model, to a hybrid representation (i.e., "huSEM") to include both undirected and directed contemporaneous effects, and then replaced the stepwise modeling with a LASSO-type regularization for a global search of the optimal sparse model. Our simulation study showed that regularized huSEM performed uniformly the best over alternative VAR representations and/or modeling approaches, with respect to accurately recovering the presence and directionality of hybrid relations and reliably removing false relations when the data are generated to have two types of contemporaneous relations. The present study to our knowledge is the first application of the recently developed regularized SEM technique to the estimation of huSEM, which points to a promising future for statistical learning in psychometric models.

Insight Into Individual Differences in Emotion Dynamics With Clustering

Studying emotion dynamics through time series models is becoming increasingly popular in the social sciences. Across individuals, dynamics can be rather heterogeneous. To enable comparisons and generalizations of dynamics across groups of individuals, one needs sophisticated tools that express the essential similarities and differences. A way to proceed is to identify subgroups of people who are characterized by qualitatively similar emotion dynamics through dynamic clustering. So far, these methods assume equal generating processes for individuals per cluster. To avoid this overly restrictive assumption, we outline a probabilistic clustering approach based on a mixture model that clusters on individuals' vector autoregressive coefficients. We evaluate the performance of the method and compare it with a nonprobabilistic method in a simulation study. The usefulness of the methods is illustrated using 366 ecological momentary assessment time series with external measures of depression and anxiety.

Detecting Task-Dependent Functional Connectivity in Group Iterative Multiple Model Estimation with Person-Specific Hemodynamic Response Functions

Introduction: Group iterative multiple model estimation (GIMME) has proven to be a reliable data-driven method to arrive at functional connectivity maps that represent associations between brain regions across time in groups and individuals. However, to date, GIMME has not been able to model time-varying task-related effects. This article introduces HRF-GIMME, an extension of GIMME that enables the modeling of the direct and modulatory effects of a task on functional magnetic resonance imaging data collected by using event-related designs. Critically, hemodynamic response function (HRF)-GIMME incorporates person-specific modeling of the HRF to accommodate known variability in onset delay and shape. Methods: After an introduction of the technical aspects of HRF-GIMME, the performance of HRF-GIMME is evaluated via both a simulation study and application to empirical data. The simulation study assesses the sensitivity and specificity of HRF-GIMME by using data simulated from one slow and two rapid event-related designs, and HRF-GIMME is then applied to two empirical data sets from similar designs to evaluate performance in recovering known neural circuitry. Results: HRF-GIMME showed high sensitivity and specificity across all simulated conditions, and it performed well in the recovery of expected relations between convolved task vectors and brain regions in both simulated and empirical data, particularly for the slow event-related design. Conclusion: Results from simulated and empirical data indicate that HRF-GIMME is a powerful new tool for obtaining directed functional connectivity maps of intrinsic and task-related connections that is able to uncover what is common across the sample as well as crucial individual-level path connections and estimates. Impact statement Group iterative multiple model estimation (GIMME) is a reliable method for creating functional connectivity maps of the connections between brain regions across time, and it is able to detect what is common across the sample and what is shared between subsets of participants, as well as individual-level path estimates. However, historically, GIMME does not model task-related effects. The novel HRF-GIMME algorithm enables the modeling of direct and modulatory task effects through individual-level estimation of the hemodynamic response function (HRF), presenting a powerful new tool for assessing task effects on functional connectivity networks in functional magnetic resonance imaging data.

Gimme’s ability to recover group-level path coefficients and individual-level path coefficients

The growing availability of intensive longitudinal data has increased psychological researchers' interest in ideographic-statistical methods that, for example, reveal the contemporaneous or lagged associations between different variables for a specific individual. However, when researchers assess several individuals, the results of such models are difficult to generalize across individuals. Researchers recently suggested an algorithm called GIMME, which allows for the identification of coefficients that exist across all individuals (group-level coefficients) or are specific to one or a subgroup of individuals (individual-level coefficients). In three simulation studies we investigated GIMME's performance in recovering group-level and individual-level coefficients. For the former, we found that GIMME performed well when the magnitude of the parameters was moderate to high and when the number of measurements was sufficiently large. However, GIMME had problems detecting individual-level coefficients or coefficients that occurred for a subset of individuals from the whole sample.

Structural equation models to estimate dynamic effective connectivity networks in resting fMRI. A comparison between individuals with Down syndrome and controls

Emerging evidence suggests that an effective or functional connectivity network does not use a static process over time but incorporates dynamic connectivity that shows changes in neuronal activity patterns. Using structural equation models (SEMs), we estimated a dynamic component of the effective network through the effects (recursive and nonrecursive) between regions of interest (ROIs), taking into account the lag 1 effect. The aim of the paper was to find the best structural equation model (SEM) to represent dynamic effective connectivity in people with Down syndrome (DS) in comparison with healthy controls. Twenty-two people with DS were registered in a functional magnetic resonance imaging (fMRI) resting-state paradigm for a period of six minutes. In addition, 22 controls, matched by age and sex, were analyzed with the same statistical approach. In both groups, we found the best global model, which included 6 ROIs within the default mode network (DMN). Connectivity patterns appeared to be different in both groups, and networks in people with DS showed more complexity and had more significant effects than networks in control participants. However, both groups had synchronous and dynamic effects associated with ROIs 3 and 4 related to the upper parietal areas in both brain hemispheres as axes of association and functional integration. It is evident that the correct classification of these groups, especially in cognitive competence, is a good initial step to propose a biomarker in network complexity studies.