BackgroundInternalising and externalising problems commonly co‐occur in childhood. Yet, few developmental models describing the structure of child psychopathology appropriately account for this comorbidity. We evaluate a model of childhood psychopathology that separates the unique and shared contribution of individual psychological symptoms into specific internalising, externalising and general psychopathology factors and assess how these general and specific factors predict long‐term outcomes concerning criminal behaviour, academic achievement and affective symptoms in three independent cohorts.MethodsData were drawn from independent birth cohorts (Avon Longitudinal Study of Parents and Children (ALSPAC), N = 11,612; Generation R, N = 7,946; Maternal Adversity, Vulnerability and Neurodevelopment (MAVAN), N = 408). Child psychopathology was assessed between 4 and 8 years using a range of diagnostic and questionnaire‐based measures, and multiple informants. First, structural equation models were used to assess the fit of hypothesised models of shared and unique components of psychopathology in all cohorts. Once the model was chosen, linear/logistic regressions were used to investigate whether these factors were associated with important outcomes such as criminal behaviour, academic achievement and well‐being from late adolescence/early adulthood.ResultsThe model that included specific factors for internalising/externalising and a general psychopathology factor capturing variance shared between symptoms regardless of their classification fits well for all of the cohorts. As hypothesised, general psychopathology factor scores were predictive of all outcomes of later functioning, while specific internalising factor scores predicted later internalising outcomes. Specific externalising factor scores, capturing variance not shared by any other psychological symptoms, were not predictive of later outcomes.ConclusionsEarly symptoms of psychopathology carry information that is syndrome‐specific as well as indicative of general vulnerability and the informant reporting on the child. The ‘general psychopathology factor' might be more relevant for long‐term outcomes than specific symptoms. These findings emphasise the importance of considering the co‐occurrence of common internalising and externalising problems in childhood when considering long‐term impact.

Score-based (denoising diffusion) generative models have recently gained a lot of success in generating realistic and diverse data. These approaches define a forward diffusion process for transforming data to noise and generate data by reversing it (thereby going from noise to data). Unfortunately, current score-based models generate data very slowly due to the sheer number of score network evaluations required by numerical SDE solvers. In this work, we aim to accelerate this process by devising a more efficient SDE solver. Existing approaches rely on the Euler-Maruyama (EM) solver, which uses a fixed step size. We found that naively replacing it with other SDE solvers fares poorly -they either result in low-quality samples or become slower than EM. To get around this issue, we carefully devise an SDE solver with adaptive step sizes tailored to score-based generative models piece by piece. Our solver requires only two score function evaluations, rarely rejects samples, and leads to high-quality samples. Our approach generates data 2 to 10 times faster than EM while achieving better or equal sample quality. For high-resolution images, our method leads to significantly higher quality samples than all other methods tested. Our SDE solver has the benefit of requiring no step size tuning. Code is available on https://github.com/AlexiaJM/score_sde_fast_sampling. † Equal contribution Preprint. Under review.

Motivated by the goal of expanding currently existing Genotype × Environment interaction (G × E) models to simultaneously include multiple genetic variants and environmental exposures in a parsimonious way, we developed a novel method to estimate the parameters in a G × E model, where G is a weighted sum of genetic variants (genetic score) and E is a weighted sum of environments (environmental score). The approach uses alternating optimization, an iterative process where the genetic score weights, the environmental score weights, and the main model parameters are estimated in turn, assuming the other parameters are constant. This technique can be used to construct relatively complex interaction models that are constrained to a particular structure, and hence contain fewer parameters. We present the model as a 2-way interaction longitudinal mixed model, for which ordinary linear regression is a special case, but it can easily be extended to be compatible with k-way interaction models and generalized linear mixed models. The model is implemented in R package) and using SAS macros Through simulations, we demonstrate the power and validity of this approach even with small sample sizes. Furthermore, we present examples from the Maternal Adversity, Vulnerability, and Neurodevelopment (MAVAN) study where we improve significantly upon already existing models using alternating optimization. (PsycINFO Database Record

Currently, two main approaches exist to distinguish differential susceptibility from diathesis-stress and vantage sensitivity in genotype × environment interaction (G×E) research: Regions of significance (RoS) and competitive-confirmatory approaches. Each is limited by their single-gene/single-environment foci given that most phenotypes are the product of multiple interacting genetic and environmental factors. We thus addressed these two concerns in a recently developed R package (LEGIT) for constructing G×E interaction models with latent genetic and environmental scores using alternating optimization. Herein we test, by means of computer simulation, diverse G×E models in the context of both single and multiple genes and environments. Results indicate that the RoS and competitive-confirmatory approaches were highly accurate when the sample size was large, whereas the latter performed better in small samples and for small effect sizes. The competitive-confirmatory approach generally had good accuracy (a) when effect size was moderate and N ≥ 500 and (b) when effect size was large and N ≥ 250, whereas RoS performed poorly. Computational tools to determine the type of G×E of multiple genes and environments are provided as extensions in our LEGIT R package.

Prenatal maternal depression and a multilocus genetic profile of two susceptibility genes implicated in the stress response were examined in an interaction model predicting negative emotionality (NE) in the first 3 years. In 179 mother-infant dyads from the Maternal Adversity, Vulnerability and Neurodevelopment cohort, prenatal depression (CES-D) was assessed at 24 to 36 weeks. The multilocus genetic profile score consisted of the number of susceptibility alleles from 5-HTTLPR (No L A (S/S, S/L G or L G /L G ) vs. any L A ) and Dopamine Receptor D4 (6-8R vs. 2-5R). NE was extracted from the IBQ-R at 3 and 6 months and the ECBQ at 18 and 36 months. Mixed and confirmatory regression analyses indicated that prenatal depression and the multilocus genetic profile interacted to predict NE from 3 to 36 months. Results were characterised by a differential susceptibility model at 3 and 6 months and by a diathesis stress model at 36 months. Negative emotionality (NE) is derived from the temperamental dimensions of sadness, distress towards limitations, fear and excessive reactions to minor changes, and reflects a generally stable tendency to show increased emotional reactivity towards negative situations (Gartstein & Rothbart, 2003;Lemery, Goldsmith, Klinnert, & Mrazek, 1999). NE is associated with the development of later problematic behaviour and psychopathology (Eisenberg et al., 2009;Fox, Henderson, Rubin, Calkins, & Schmidt, 2001;Hyde, Mezulis, & Abramson, 2008). For example, fearful temperament is associated with childhood anxiety disorders (Degnan, Alma, & Fox, 2010;Goldsmith & Lemery, 2000), while NE is associated with depression (Phillips, Lonigan, & Driscoll, 2002) and maladjustment (Eisenberg et al., 2009). Understanding early influences of NE on socio-emotional development (Davis, Glynn, Schetter, Hobel, Chicz-demet, & Sandman, 2007;Davis, Snidman, Glynn, Dunkel Schetter, & Sandman, 2004;Hayden et al., 2010;Hayden et al., 2007) could inform efforts at prevention and early intervention. Recent contradictory findings about the role of genetic and prenatal adversity (Braithwaite et al., 2013;, suggest the need for replication (Duncan, 2013) and modeling of genetic risk with multiple genes (Plomin, 2013). Accordingly, we present the findings from a study of the development of NE from 3 to 36 months of age from the interaction of prenatal maternal depression and a multilocus genetic profile. The Role of Prenatal Maternal StressPrenatal maternal stress, measured in diverse ways is associated with NE (Glover, 2011;O'Connor, Heron, & Glover, 2002a). For example, higher prenatal maternal cortisol is associated with fussier behaviour, more negative facial expressions, crying, as well as higher NE at 7 weeks of age (de Weerth, Hees, & Buitelaar, 2003). Symptoms of prenatal maternal anxiety and depression, loosely associated with the stress response predict behavioural reactivity at 4 months of age (Davis et al., 2004) and behavioural/emotional problems at 4 years of age (O'Connor, et al., 2002a). While most studi...

Objective: Few studies have attempted to identify how distinct dimensions of maternal prenatal affective symptoms relate to offspring psychopathology. We defined latent dimensions of women's prenatal affective symptoms and pregnancy-specific worries to examine their association with early offspring psychopathology in three prenatal cohorts. Method: Data were used from three cohorts of the DREAM-BIG consortium: Avon Longitudinal Study of Parents and Children (ALSPAC [N ¼ 12,515]), Generation R (N ¼ 6,803), and the Canadian prenatal cohort Maternal Adversity, Vulnerability, and Neurodevelopment (MAVAN [N ¼ 578]). Maternal prenatal affective symptoms and pregnancy-specific worries were assessed using different measures in each cohort. Through confirmatory factor analyses, we determined whether comparable latent dimensions of prenatal maternal affective symptoms existed across the cohorts. We used structural equation models to examine cohort-specific associations between these dimensions and offspring psychopathology at 4 to 8 years of age (general psychopathology, specific internalizing and externalizing previously derived using confirmatory factor analyses). Cohort-based estimates were meta-analyzed using inverse variance-weighing. Results: Four prenatal maternal factors were similar in all cohorts: a general affective symptoms factor and three specific factors-an anxiety/depression factor, a somatic factor, and a pregnancy-specific worries factor. In meta-analyses, both the general affective symptoms factor and pregnancy-specific worries factor were independently associated with offspring general psychopathology. The general affective symptoms factor was further associated with offspring specific internalizing problems. There were no associations with specific externalizing problems. Conclusion: These replicated findings of independent and adverse effects for prenatal general affective symptoms and pregnancy-specific worries on child mental health support the need for specific interventions in pregnancy.

Denoising score matching with Annealed Langevin Sampling (DSM-ALS) is a recent approach to generative modeling. Despite the convincing visual quality of samples, this method appears to perform worse than Generative Adversarial Networks (GANs) under the Fréchet Inception Distance, a popular metric for generative models. We show that this apparent gap vanishes when denoising the final Langevin samples using the score network. In addition, we propose two improvements to DSM-ALS: 1) Consistent Annealed Sampling as a more stable alternative to Annealed Langevin Sampling, and 2) a hybrid training formulation, composed of both denoising score matching and adversarial objectives. By combining both of these techniques and exploring different network architectures, we elevate score matching methods and obtain results competitive with state-of-the-art image generation on CIFAR-10.

Purpose of review Understanding variability in developmental outcomes following exposure to early life adversity (ELA) has been an area of increasing interest in psychiatry, as resilient outcomes are just as prevalent as negative ones. However, resilient individuals are understudied in most cohorts and even when studied, resilience is typically defined as an absence of psychopathology. This review examines current approaches to resilience and proposes more comprehensive and objective ways of defining resilience. Recent findings Of the 36 studies reviewed, the most commonly used measure was the Strengths and Difficulties Questionnaire (n = 6), followed by the Child Behavior Checklist (n = 5), the Resilience Scale for Chinese Adolescents (n = 5), the Rosenberg Self-Esteem Scale (n = 4), and the Child and Youth Resilience Scale (n = 3). Summary This review reveals that studies tend to rely on self-report methods to capture resilience which poses some challenges. We propose a complementary measure of child resilience that relies on more proactive behavioral and observational indicators; some of our preliminary findings are presented. Additionally, concerns about the way ELA is characterized as well as the influence of genetics on resilient outcomes prompts further considerations about how to proceed with resiliency research.