Improving the Walktrap Algorithm Using K-Means Clustering.

The walktrap algorithm is one of the most popular community-detection methods in psychological research. Several simulation studies have shown that it is often effective at determining the correct number of communities and assigning items to their proper community. Nevertheless, it is important to recognize that the walktrap algorithm relies on hierarchical clustering because it was originally developed for networks much larger than those encountered in psychological research. In this paper, we present and demonstrate a computational alternative to the hierarchical algorithm that is conceptually easier to understand. More importantly, we show that better solutions to the sum-of-squares optimization problem that is heuristically tackled by hierarchical clustering in the walktrap algorithm can often be obtained using exact or approximate methods for K-means clustering. Three simulation studies and analyses of empirical networks were completed to assess the impact of better sum-of-squares solutions.

Methods for quantifying the heterogeneity of psychopathology.

OBJECTIVES: Specifiers for a major depressive disorder (MDE) are supposed to reduce diagnostic heterogeneity. However, recent literature challenges the idea that the atypical and melancholic specifiers identify more homogenous or coherent subgroups. We introduce the usage of distance metrics to characterize symptom heterogeneity. We attempt to replicate prior findings and explore whether symptom heterogeneity is reduced using specifier subgroups. METHODS: We used data derived from the National Epidemiological Survey on Alcohol and Related Conditions (NESARC Wave I; N = 5,749) and the Sequenced Treatment Alternatives to Relieve Depression study (STAR*D; N = 2,498). We computed Hamming and Manhattan distances from study participants' unique symptom profiles. Distances were standardized from 0-1 and compared by their within- and between-group similarities to their non-specifier counterparts for the melancholic and atypical specifiers. RESULTS: There was no evidence of statistically significant differences in heterogeneity for specifier (i.e., melancholic or atypical) vs. non-specifier designations (i.e., non-melancholic vs. non-atypical). CONCLUSION: Replicating prior work, melancholic and atypical depression specifiers appear to have limited utility in reducing heterogeneity. The current study does not support the claim that specifiers create more coherent subgroups as operationalized by similarity in the number of symptoms and their severity. Distance metrics are useful for quantifying symptom heterogeneity.

Psychopathology as dynamic markers of alcohol initiation across development: A three-year longitudinal examination.

Sipping, an early form of alcohol initiation, is associated with aspects of psychopathology and personality that reflect long-term risk for harmful alcohol use. In the Adolescent Brain and Cognitive Development cohort (N = 11,872), sipping by age 9-10 was concurrently associated with impulsivity, other aspects of externalizing, and prodromal schizophrenia symptoms. Still, these associations were cross-sectional in nature, leaving open the possibility that these features of psychopathology and personality might not reflect long-term risk for alcohol consumption and related harm across development. Here, we attempted to replicate baseline concurrent associations across three waves of data to extend concurrent associations to prospective ones. Most cross-sectional associations replicated across waves, such that impulsivity, other aspects of externalizing, reward sensitivity (e.g., surgency, sensation seeking), and prodromal schizophrenia symptoms were associated with increased odds of having sipped alcohol by the age of 12. Nevertheless, not all concurrent associations replicated prospectively; impulsigenic features did not reflect long-term risk for sipping. Thus, some psychopathology features appeared to reflect stable risk factors, whereas others appeared to reflect state-dependent risk factors. All told, sipping might not reflect long-term risk for harmful alcohol use, and the nature of sipping may change across development.

Focusing Narrowly on Model Fit in Factor Analysis Can Mask Construct Heterogeneity and Model Misspecification: Applied Demonstrations across Sample and Assessment Types.

This study builds upon research indicating that focusing narrowly on model fit when evaluating factor analytic models can lead to problematic inferences regarding the nature of item sets, as well as how models should be applied to inform measure development and validation. To advance research in this area, we present concrete examples relevant to researchers in clinical, personality, and related subfields highlighting two specific scenarios when an overreliance on model fit may be problematic. Specifically, we present data analytic examples showing that focusing narrowly on model fit may lead to (a) incorrect conclusions that heterogeneous item sets reflect narrower homogeneous constructs and (b) the retention of potentially problematic items when developing assessment measures. We use both interview data from adult outpatients (N = 2,149) and self-report data from adults recruited online (N = 547) to demonstrate the importance of these issues across sample types and assessment methods. Following demonstrations with these data, we make recommendations focusing on how other model characteristics (e.g., factor loading patterns; carefully considering the content and nature of factor indicators) should be considered in addition to information provided by model fit indices when evaluating factor analytic models.

On maximization of the modularity index in network psychometrics.

The modularity index (Q) is an important criterion for many community detection heuristics used in network psychometrics and its subareas (e.g., exploratory graph analysis). Some heuristics seek to directly maximize Q, whereas others, such as the walktrap algorithm, only use the modularity index post hoc to determine the number of communities. Researchers in network psychometrics have typically not employed methods that are guaranteed to find a partition that maximizes Q, perhaps because of the complexity of the underlying mathematical programming problem. In this paper, for networks of the size commonly encountered in network psychometrics, we explore the utility of finding the partition that maximizes Q via formulation and solution of a clique partitioning problem (CPP). A key benefit of the CPP is that the number of communities is naturally determined by its solution and, therefore, need not be prespecified in advance. The results of two simulation studies comparing maximization of Q to two other methods that seek to maximize modularity (fast greedy and Louvain), as well as one popular method that does not (walktrap algorithm), provide interesting insights as to the relative performances of the methods with respect to identification of the correct number of communities and the recovery of underlying community structure.

A comparison of logistic regression methods for Ising model estimation.

The Ising model has received significant attention in network psychometrics during the past decade. A popular estimation procedure is IsingFit, which uses nodewise l1-regularized logistic regression along with the extended Bayesian information criterion to establish the edge weights for the network. In this paper, we report the results of a simulation study comparing IsingFit to two alternative approaches: (1) a nonregularized nodewise stepwise logistic regression method, and (2) a recently proposed global l1-regularized logistic regression method that estimates all edge weights in a single stage, thus circumventing the need for nodewise estimation. MATLAB scripts for the methods are provided as supplemental material. The global l1-regularized logistic regression method generally provided greater accuracy and sensitivity than IsingFit, at the expense of lower specificity and much greater computation time. The stepwise approach showed considerable promise. Relative to the l1-regularized approaches, the stepwise method provided better average specificity for all experimental conditions, as well as comparable accuracy and sensitivity at the largest sample size.

Alcohol use disorder conceptualizations and diagnoses reflect their sociopolitical context.

The present paper highlights how alcohol use disorder (AUD) conceptualizations and resulting diagnostic criteria have evolved over time in correspondence with interconnected sociopolitical influences in the United States. We highlight four illustrative examples of how DSM-defined alcoholism, abuse/dependence, and AUD have been influenced by sociopolitical factors. In doing so, we emphasize the importance of recognizing and understanding such sociopolitical factors in the application of AUD diagnoses. Last, we offer a roadmap to direct the process of future efforts toward the improved diagnosis of AUD, with an emphasis on pursuing falsifiability, acknowledging researchers' assumptions about human behavior, and collaborating across subfields. Such efforts that center the numerous mechanisms and functions of behavior, rather than signs or symptoms, have the potential to minimize sociopolitical influences in the development of diagnostic criteria and maximize the treatment utility of diagnoses.

Sociodemographic differences in youth alcohol sipping's nomological network.

BACKGROUND: Previous research has established that certain features of personality (e.g., impulsivity), psychopathology (e.g., impulsivity, mood disorder, thought disorder), and contextual factors (e.g., parenting, parental alcohol use) are associated with an increased likelihood of having sipped alcohol in youth, and substance involvement and problems in adolescence and adulthood. What is less clear from the existing literature is whether well-established risk factors of substance use are consistent across sociodemographic characteristics (i.e., gender, race/ethnicity, religious affiliation, income, parental education). METHODS: We used a large, community sample of 9- and 10-year-olds (N = 11,872; 53% female) to examine whether various sociodemographic characteristics moderate the associations between sipping behavior and its various well-established correlates (e.g., impulsivity, behavioral inhibition and activation, psychopathology, parenting, and family conflict). RESULTS: There were small mean level differences in sipping across sociodemographic characteristics. Across sociodemographic characteristics, however, sipping was fairly uniformly associated with youth-reported impulsivity, behavioral activation, prodromal psychosis symptoms, mood and externalizing disorder diagnoses, family environment, and parental alcohol consumption indices. Effects were sometimes slightly more pronounced among groups for which alcohol consumption is relatively nonnormative: Sipping among female youth was slightly more associated with thought disorder psychopathology than among male youth (D = 0.07), and was slightly more associated with some aspects of psychopathology and impulsivity for Black youth than White and Hispanic youth (Ds were 0.07 and 0.09). CONCLUSIONS: Broadly, our findings suggest that the psychosocial correlates of precocious alcohol use are relatively consistent across sociodemographic factors.

The distinction between symptoms and traits in the Hierarchical Taxonomy of Psychopathology (HiTOP).

The Hierarchical Taxonomy of Psychopathology (HiTOP) is an empirically and quantitatively derived dimensional classification system designed to describe the features of psychopathology and, ultimately, to replace categorical nosologies. Among the constructs that HiTOP organizes are "symptom components" and "maladaptive traits," but past HiTOP publications have not fully explicated the distinction between symptoms and traits. We propose working definitions of symptoms and traits and explore challenges, exceptions, and remaining questions. Specifically, we propose that the only systematic difference between symptoms and traits in HiTOP is one of time frame. Maladaptive traits are dispositional constructs that describe persistent tendencies to manifest features of psychopathology, whereas symptoms are features of psychopathology as they are manifest during any specific time period (from moments to days to months). This has the consequence that almost every HiTOP dimension, at any level of the hierarchy, can be assessed as either a trait or a symptom dimension, by adjusting the framing of the assessment. We discuss the implications of these definitions for causal models of the relations between symptoms and traits and for distinctions between psychopathology, normal personality variation, and dysfunction.

Meaningful associations in the adolescent brain cognitive development study.

The Adolescent Brain Cognitive Development (ABCD) Study is the largest single-cohort prospective longitudinal study of neurodevelopment and children's health in the United States. A cohort of n = 11,880 children aged 9-10 years (and their parents/guardians) were recruited across 22 sites and are being followed with in-person visits on an annual basis for at least 10 years. The study approximates the US population on several key sociodemographic variables, including sex, race, ethnicity, household income, and parental education. Data collected include assessments of health, mental health, substance use, culture and environment and neurocognition, as well as geocoded exposures, structural and functional magnetic resonance imaging (MRI), and whole-genome genotyping. Here, we describe the ABCD Study aims and design, as well as issues surrounding estimation of meaningful associations using its data, including population inferences, hypothesis testing, power and precision, control of covariates, interpretation of associations, and recommended best practices for reproducible research, analytical procedures and reporting of results.

Incipient alcohol use in childhood: Early alcohol sipping and its relations with psychopathology and personality.

Prior research has shown that sipping of alcohol begins to emerge during childhood and is potentially etiologically significant for later substance use problems. Using a large, community sample of 9- and 10-year-olds (N = 11,872; 53% female), we examined individual differences in precocious alcohol use in the form of alcohol sipping. We focused explicitly on features that are robust and well-demonstrated correlates of, and antecedents to, alcohol excess and related problems later in the lifespan, including youth- and parent-reported externalizing traits (i.e., impulsivity, behavioral inhibition and activation) and psychopathology. Seventeen percent of the sample reported sipping alcohol outside of a religiously sanctioned activity by age 9 or 10. Several aspects of psychopathology and personality emerged as small but reliable correlates of sipping. Nonreligious sipping was related to youth-reported impulsigenic traits, aspects of behavioral activation, prodromal psychotic-like symptoms, and mood disorder diagnoses, as well as parent-reported externalizing disorder diagnoses. Religious sipping was unexpectedly associated with certain aspects of impulsivity. Together, our findings point to the potential importance of impulsivity and other transdiagnostic indicators of psychopathology (e.g., emotion dysregulation, novelty seeking) in the earliest forms of drinking behavior.

Psychiatric Comorbidity as a Function of Severity: DSM-5 Alcohol Use Disorder and HiTOP Classification of Mental Disorders.

BACKGROUND: Understanding the comorbidity of alcohol use disorder (AUD) and other psychiatric diagnoses has been a long-standing interest of researchers and mental health professionals. Comorbidity is often examined via the diagnostic co-occurrence of discrete, categorical diagnoses, which is incongruent with increasingly supported dimensional approaches of psychiatric classification and diagnosis, and for AUD more specifically. The present study examined associations between DSM-5 AUD and psychiatric symptoms of other DSM-IV and DSM-5 disorders categorically, and dimensionally organized according to the Hierarchical Taxonomy of Psychopathology (HiTOP) spectra (e.g., Internalizing, Disinhibited Externalizing). METHODS: The comorbidity of AUD with other psychological disorders was examined in 2 independent nationally representative samples of past-year drinkers via an initial examination in the National Epidemiological Survey on Alcohol and Related Conditions (NESARC) Wave 2 and replicated in NESARC-III. RESULTS: Analyses focusing on psychopathology symptom counts organized by spectra demonstrated that greater AUD severity was associated with a higher number of symptoms across HiTOP spectra. Traditional categorical analyses also demonstrated increasing prevalence as a monotonic function of DSM-5 AUD severity gradients. CONCLUSIONS: This study indicates that AUD and other psychiatric disorder comorbidity implies increased presence of multiple forms of psychopathology with a corresponding increased number of symptoms across hierarchical spectra. Greater AUD severity increases the likelihood of other psychopathology and, when present, "more severe" presentations. That is, on average, a given disorder (e.g., depression) is more severe when copresenting with an AUD, and increases in severity along with the AUD.

A Longitudinal and Multidimensional Examination of the Associations Between Temperament and Self-Restraint During Toddlerhood.

Developing self-restraint, or the inhibition of behavior in response to a prohibition, is an important process during toddlerhood. The objective of this study was to gain a better understanding of individual differences in the development of self-restraint during toddlerhood by examining stable elements and growth of temperament (i.e., attentional control, behavioral inhibition, negative emotionality), general intelligence, and self-restraint. Participants were 412 same-sex twin pairs (approximately 90% European American) from predominately middle-class households in Colorado. Data were collected at 14, 20, 24, and 36 months. Results indicated that higher behavioral inhibition, attentional control, and intelligence were independently associated with better self-restraint, whereas higher negative emotionality was an independent predictor of lower self-restraint. The associations between temperament and self-restraint generally appeared to be stable from 14 to 36 months.

Foundations of Arrogance: A Broad Survey and Framework for Research.

We consider the topic of arrogance from a cross-disciplinary viewpoint. To stimulate further research, we suggest three types of arrogance (individual, comparative, and antagonistic) and six components contributing to them, each logically related to the next. The components progress from imperfect knowledge and abilities to an unrealistic assessment of them, an unwarranted attitude of superiority over other people, and related derisive behavior. Although each component presumably is present to some degree when the next one operates, causality might flow between components in either direction. The classification of components of arrogance should reduce miscommunication among researchers, as the relevant concepts and mechanisms span cognitive, motivational, social, and clinical domains and literatures. Arrogance is an important concept warranting further study for both theoretical and practical reasons, in both psychopathology and normal social interaction. Everyone seems to have qualities of arrogance to some degree, and we consider the importance of arrogance on a spectrum. We contend that humankind can benefit from a better understanding of the cognitive limitations and motivational biases that, operating together, appear to contribute to arrogance. We bring together information and questions that might lead to an invigorating increase in the rate and quality of cross-disciplinary research on arrogance.

Do psychopathic birds of a feather flock together? Psychopathic personality traits and romantic preferences.

OBJECTIVE: The goal of the present studies was to investigate whether people are especially attracted to psychopathic traits, and whether there are individual differences in such attraction. METHOD: Female undergraduates (N = 270; Mage = 19; 57% White, 20% Asian, 8% Black) and female and male community members (N = 426; Mage = 37; 56% female; 81% Caucasian, 10% African American, 4% Asian) reported on their own personality and constructed their ideal mate for a dating, short-term, and long-term relationship from a list of 70 characteristics drawn from well-validated criteria for psychopathic personality and diagnostic criteria for DSM-5 personality disorders (PDs). RESULTS: Across both studies, absolute romantic preferences for psychopathic traits collapsed across time point were low on average, but higher than those for most all other PDs. In addition, they were higher for Factor 1 (i.e., interpersonal/affective) as opposed to Factor 2 (i.e., impulsive, antisocial) psychopathy traits. Participants with marked PD features, including Factor 2 psychopathy traits, were more inclined than others to endorse a preference for psychopathic males. CONCLUSIONS: Relative attraction to psychopathic males and observed homophily may be avenues through which psychopathic traits persist in the population across time.

The Association Between Toddlerhood Self-Control and Later Externalizing Problems.

Lower self-control is a significant correlate or predictor of a wide range of adult outcomes, and this association may be due to more general tendencies toward childhood externalizing problems. The present study examined the association between toddlerhood self-control expressed within a "don't" compliance task (at 14-36 months) and later externalizing problems (parent-reported externalizing problems from age 4 to 12 years, teacher-reported externalizing problems from age 7 to 12 years, and self-reported conduct disorder symptoms at age 17 years) in a longitudinal, genetically informative study. The slope of self-control, but not its intercept, predicted later teacher-reported, but not parent- or self-reported, externalizing problems. That is, increase in self-control during toddlerhood was associated with lower levels of later teacher-reported externalizing problems. The slope of self-control was no longer a significant predictor of teacher-reported externalizing problems after controlling for observed disregard for others, a robust predictor of externalizing problems. Thus, the hypothesis that self-control is the primary predictor of externalizing problems was not supported. Results from genetic analyses suggested that the covariance between the slope of self-control and teacher-reported externalizing problems is due to both genetic and shared environmental influences.

Examining the factor structure of narrative meaning-making for stressful events and relations with psychological distress.

This study examined the underlying factor structure of 15 narrative meaning-making indices for narratives of stressful events, and explored the incremental validity of the narrative factor solution over and above general personality traits in predicting various indices of psychological well-being. Two-hundred and twenty four undergraduates (Mage = 19.2 years, SDage = 2.1; 114 males and 110 females; 67.6% Caucasian, 12.0% East Asian, 7.6% African-American, 4.0% South Asian, 2.2% Hispanic, and 6.7% as mixed or Other origin) wrote about the most traumatic experience in their life, and completed a series of psychological questionnaires. The narratives were coded in 15 ways theoretically derived from the narrative meaning-making literature. A series of exploratory structural equation models indicated that a four-factor solution best approximated the data. The four factors were: positive processing, negative processing, integrative meaning, and structure. All four factors related differentially to indices of well-being over and above traits. There appear to be four distinct, but related, factors of narrative meaning-making for memories of stressful events, which shed light on the nuanced relations with well-being.

Psychopathy Deconstructed and Reconstructed: Identifying and Assembling the Personality Building Blocks of Cleckley's Chimera.

The psychopathy field has long been beset by confusion and contention regarding the boundaries and features of this chimerical condition. We propose that this disagreement stems largely from the historical separation between psychopathy and basic personality psychology. Using findings from a meta-analysis of the correlations between the Psychopathy Checklist-Revised (PCL-R) and normal-range personality traits as a launching point, we (a) deconstruct widely used measures of psychopathy into their constituent subdimensions and (b) examine the associations of these subdimensions with higher-order and lower-order personality dimensions drawn from the Big Five and Big Three frameworks. Our review of the adult psychopathy literature reveals broad agreement that psychopathy measures are imbued with low Agreeableness and low Conscientiousness. Nevertheless, substantial disagreement revolves around the place of largely adaptive features, especially high agentic Extraversion, low Neuroticism, and high Openness, within the psychopathy construct. We propose that ongoing debates regarding the nature and boundaries of psychopathy reflect a focus on two differing operationalizations of this condition, each of which reflects a different "species" of individual.

Testing alternative hypotheses regarding the association between behavioral inhibition and language development in toddlerhood.

Studies have reported an inverse association between language development and behavioral inhibition or shyness across childhood, but the direction of this association remains unclear. This study tested alternative hypotheses regarding this association in a large sample of toddlers. Data on behavioral inhibition and expressive and receptive language abilities were collected from 816 twins at ages 14, 20, and 24 months. Growth and regression models were fit to the data to assess the longitudinal associations between behavioral inhibition and language development from 14 to 24 months. Overall, there were significant associations between behavioral inhibition and expressive language, and minimal associations with receptive language, indicating that the association is better explained by reticence to respond rather than deficient language development.