Background and aims The DSM-5 includes criteria for diagnosing Internet gaming disorder (IGD) that are adapted from substance abuse and widely used in research and clinical contexts, although evidence supporting their validity remains scarce. This study compared online gamers who do or do not endorse IGD criteria regarding self-control-related abilities (impulsivity, inhibitory control, and decision-making), considered the hallmarks of addictive behaviors. Method A double approach was adopted to distinguish pathological from recreational gamers: The first is the classic DSM-5 approach (≥5 criteria required to endorse the IGD diagnosis), and the second consists in using latent class analysis (LCA) for IGD criteria to distinguish gamers’ subgroups. We computed comparisons separately for each approach. Ninety-seven volunteer gamers from the community were recruited. Self-reported questionnaires were used to measure demographic- and game-related characteristics, problematic online gaming (with the Problematic Online Gaming Questionnaire), impulsivity (with the UPPS-P Impulsive Behavior Scale), and depression (with the Beck Depression Inventory-II). Experimental tasks were used to measure inhibitory control (Hybrid-Stop Task) and decision-making abilities (Game of Dice Task). Results Thirty-two participants met IGD criteria (33% of the sample), whereas LCA identified two groups of gamers [pathological (35%) and recreational]. Comparisons that used both approaches (DSM-5 and LCA) failed to identify significant differences regarding all constructs except for variables related to actual or problematic gaming behaviors. Discussion The validity of IGD criteria is questioned, mostly with respect to their relevance in distinguishing high engagement from pathological involvement in video games.
In this Master Thesis, we have analytically derived and numerically implemented three estimators of the Prediction Divergence Criterion (Avella-Medina et al., working paper) for Model Selection within the logistic regression framework. After the validation of these estimators by means of simulations, we have performed Model Selection both when the order of the variables was known in advance and when the order was correct but decided by an already existing algorithm, namely the binary lasso (Friedman et al., 2010). Finally we have produced evidences of the good performance of two of these estimators, one derived from the L2 norm error measure and the other from the binomial deviance, respectively in highly and moderately correlated settings. They have been proven better, to the extension of the simulation study, than the defaults methods, based on 10-fold Cross Validation, currently available in the glmnet(Friedman et al., 2017) R package.
In this thesis I investigate the validity and the accuracy properties of the posterior quantiles in Bayesian statistics when replacing the parametric likelihood with the Cressie-Read empirical likelihoods based on a set of unbiased M-estimating equations. At first order I study the validity of the empirical posterior distribution derived from the pseudo-likelihood constructed with profiled weights and estimated at a minimum distance from the empirical distribution in the Cressie-Read family of divergences, indexed by γ. The bias in coverage of the resulting empirical posterior quantile is inversely proportional to the asymptotic efficiency of the estimator corresponding to the set of M-estimating functions. By comparing different members of the Cressie-Read family of empirical likelihoods for models in the exponential family, I establish a hierarchy in the accuracy of the quantile function of the empirical posterior distribution depending on the index parameter γ.
Background The Psychiatric arm of the population-based CoLaus study (PsyCoLaus) is designed to: 1) establish the prevalence of threshold and subthreshold psychiatric syndromes in the 35 to 66 year-old population of the city of Lausanne (Switzerland); 2) test the validity of postulated definitions for subthreshold mood and anxiety syndromes; 3) determine the associations between psychiatric disorders, personality traits and cardiovascular diseases (CVD), 4) identify genetic variants that can modify the risk for psychiatric disorders and determine whether genetic risk factors are shared between psychiatric disorders and CVD. This paper presents the method as well as sociodemographic and somatic characteristics of the sample. Methods All 35 to 66 year-old persons previously selected for the population-based CoLaus survey on risk factors for CVD were asked to participate in a substudy assessing psychiatric conditions. This investigation included the Diagnostic Interview for Genetic Studies to elicit diagnostic criteria for threshold disorders according to DSM-IV and algorithmically defined subthreshold syndromes. Complementary information was collected on potential risk and protective factors for psychiatric disorders, migraine and on the morbidity of first-degree relatives, whereas the collection of DNA and plasma samples was already part of the original CoLaus survey. Results A total of 3,691 individuals completed the psychiatric evaluation (67% participation). The gender distribution of the sample did not differ significantly from that of the general population in the same age range. Although the youngest 5-year band of the cohort was underrepresented and the oldest 5-year band overrepresented, participants of PsyCoLaus and individuals who refused to participate revealed comparable scores on the General Health Questionnaire, a self-rating instrument completed at the somatic exam. Conclusion Despite limitations resulting from the relatively low participation in the context of a comprehensive and time-consuming investigation, the PsyCoLaus study should significantly contribute to the current understanding of psychiatric disorders and comorbid somatic conditions by: 1) establishing the clinical relevance of specific psychiatric syndromes below the DSM-IV threshold; 2) determining comorbidity between risk factors for CVD and psychiatric disorders; 3) assessing genetic variants associated with common psychiatric disorders and 4) identifying DNA markers shared between CVD and psychiatric disorders.