The maturation and consolidation of executive functions, including cognitive flexibility, attentional control, goal-setting, and information processing, continues throughout adolescence. Cognitive vulnerabilities to depression, such as rumination on negative affect, negative cognitive style, and hopelessness, also emerge as stable risk-factors for depression during this time. Emerging evidence suggests these vulnerabilities may be associated with alterations in executive functioning, and with cognitive maturation. The current study explores the association between trajectories of executive development and cognitive vulnerabilities to depression using a person-centered characterization of latent classes of growth trajectories. Classes of adolescent cognitive development in working memory, selective attention, sustained attention, switching, and divided attention, were derived, and class associations with cognitive vulnerabilities were probed. The results showed that most executive domains have a normative majority with typical growth and low levels of cognitive vulnerability. Minority classes, representing atypical growth, were differentially related to cognitive vulnerability. Contrary to hypotheses, better cognitive development was generally associated with higher levels of cognitive vulnerability, specifically internal, stable, and self-worth dimensions of negative cognitive style. Several exceptions included classes whose trajectory suggested developmental regression; consistent with hypotheses, these classes also demonstrated higher levels of negative cognitive style. Results support a model in which cognitive development scaffolds the maturation of negative cognitive style. / Psychology

Due to difficulties in elucidating neurobiological aspects of psychological disorders, the National Institute of Mental Health (NIMH) created the Research Domain Criteria (RDoC), which encourages novel conceptualizations of the relationship between neurobiological circuitry and clinical difficulties. This approach is markedly different from the Diagnostic and Statistical Manual of Mental Disorders (DSM) based approach that has dominated clinical research to date. Thus, RDoC necessitates exploration of novel experimental and statistical approaches. Fear learning paradigms represent a promising methodology for elucidating connections between acute threat (“fear”) circuitry and fear-related clinical difficulties. However, traditional analytical approaches rely on central tendency statistics, which are tethered to a priori categories and assume homogeneity within groups. Growth Mixture Modeling (GMM) methods such as Latent Class Growth Analysis (LCGA) may be uniquely suited for examining fear learning phenotypes. However, just three extant studies have applied GMM to fear learning and only one did so in a human population. Thus, the degree to which classes identified in known studies represent characteristics of the general population and to which GMM methodology is applicable across populations and paradigms is unclear. This preliminary study applied LCGA to a fear learning lab study in an attempt to identify heterogeneity in fear learning patterns based on a posteriori classification. The findings of this investigation may inform efforts to move toward a trans-diagnostic conceptualization of fear learning. Consistent with the goals laid out in RDoC, explication of fear learning phenotypes may eventually provide critical information needed to spur innovation in psychotherapeutic and psychopharmacological treatment. / Master of Science / To date, most clinical psychology research has been based on the Diagnostic and Statistical Manual of Mental Disorders (DSM), which is a catalog of mental health disorders that was originally designed to facilitate communication among clinicians. Many experts contend that this approach has hampered progress in the field of biological clinical psychology research. Thus, the National Institute of Mental Health (NIMH) created a new template for biological clinical psychology research called the Research Domain Criteria (RDoC). Since RDoC calls for a complete overhaul in the conceptualization of clinical dysfunction, this approach requires statistical and experimental innovation. One traditional experimental approach that may be helpful in understanding the RDoC topic of acute threat (“fear”) is called Pavlovian Fear Learning (PFL). However, traditional PFL studies have utilized statistical methods that are based on comparing group averages and require researchers to determine groups of interest based on theory before the study begins. This is problematic because RDoC calls for research that begins with evidence rather than theory. Growth Mixture Modeling (GMM) is a statistical methodology that may allow researchers to analyze fear learning data without having to begin with theoretically determined categories such as DSM disorders. However, little research has tested how well this approach would work. This study is just the second to apply a GMM approach to a human PFL study. The findings from this investigation may inform efforts to develop a statistical technique that is well suited for RDoCian research and may also spur innovation in psychotherapeutic and psychopharmacological treatment.

The Outcome Questionnaire (OQ-45; Lambert et al., 1996) has been widely employed as a psychotherapy outcome monitoring measure following research findings that support various aspects of its validity and sensitivity to change. Despite its broad usage in both clinical and research settings, some of its psychometric properties are not definite. The three subscales of the OQ-45 are designed to measure three distinct, but related, aspects of psychological functioning. However, neither the one- nor three-factor models have been supported by previous research. Likewise, the results of the current study supported neither of those factor structures. It was suspected that heterogeneity in data might have led to the lack of the confirmatory factor analysis model fit. Therefore, factor mixture modeling (FMM), a combination of confirmatory factor analysis and latent class analysis, was employed to investigate potential heterogeneity of the data. Among the series of factor mixture models with varying numbers of classes that were fitted, the two-class, unconditional FMM based on the revised three-factor solution was decided to best describe the data under analysis. Although three covariates of clinical status, sex, and race were selected as known sources of heterogeneity and incorporated into the FMMs (i.e., conditional model), the findings were contradictory to expectations. The implications of these findings in counseling were discussed in terms of aggregating OQ-45 scores and its score interpretation. Furthermore, this study demonstrates the process involved and dilemmas encountered in choosing the best fitting FMM. There is currently no criterion for assessing individual model fit. Instead, models’ fit are compared using various information criteria (IC). And, as was found in the current study, these ICs are frequently contradictory. Thus, the process of identifying the best fitting model cannot rest solely on fit indices but must also depend on interpretation of models and consideration of the ultimate use of the results. In the current study, consideration of transition matrices and the pattern of latent means across classes contributed as much to model selection as fit index interpretation. / text

Doctor of Philosophy / Department of Family Studies and Human Services / Jared A. Durtschi and Sandra M. Stith / Early binge drinking (i.e., five or more drinks on a single occasion) is associated with a greater risk of later substance abuse or dependence, and other non-alcohol related problems in adulthood, (e.g., adult civil or criminal convictions). Identifying alcohol use trajectories has mainly been limited to within single developmental periods (i.e., adolescence or emerging adulthood) or between developmental periods up until around the legal drinking age. Using N = 1,864 adolescents from the National Longitudinal Study of Adolescent Health (Add Health) dataset, this paper sought to identify trajectories of binge drinking beginning in adolescence and into adulthood using growth mixture modeling. Family factors (e.g., parent-child communication, shared activities, connectedness, and parental control) were used to predict the various trajectories. Two class trajectories were identified, a low initial-escalating group (87%), and a high initial-deescalating group (13%). Being male and having more close friends using alcohol were predictive of a greater likelihood of being in the high initial-deescalating group. Results can inform therapeutic interventions in an effort to affect an adolescent’s trajectory of use and reduce the risk of long-term heavy alcohol use.

<p>This thesis focuses on the development of ABC methods for statistical modeling in complex dynamic systems. Motivated by real applications in biology, I propose computational strategies for Bayesian inference in contexts where standard Monte Carlo methods cannot be directly applied due to the high complexity of the dynamic model and/or data limitations.</p><p> Chapter 2 focuses on stochastic bionetwork models applied to data generated from the marginal distribution of a few network nodes at snapshots in time. I present a Bayesian computational strategy, coupled with an approach to summarizing and numerically characterizing biological phenotypes that are represented in terms of the resulting sample distributions of cellular markers. ABC and mixture modeling are used to define the approach to linking mechanistic mathematical models of network dynamics to snapshot data, using a toggle switch example integrating simulated and real data as context. </p><p> Chapter 3 focuses on the application of the methodology presented in Chapter 2 to the Myc/Rb/E2F network. This network involves a relatively high number of parameters and stochastic equations in the model specification and, thus, is substantially more complex than the toggle switch example. The analysis of the Myc/Rb/E2F network is performed with simulated and real data. I demonstrate that the proposed method can indicate which parameters can be learned about using the marginal data. </p><p> In Chapter 4, I present an ABC SMC method that uses data-based adaptive weights. This easily implemented and computationally trivial extension of ABC SMC can substantially improve acceptance rates. This is demonstrated through a series of examples with simulated and real data, including the toggle switch example. Theoretical justification is also provided to explain why this method is expected to improve the effectiveness of ABC SMC.</p><p> In Chapter 5, I present an integrated Bayesian computational strategy for fitting complex dynamic models to sparse time-series data. This is applied to experimental data from an immunization response study with Indian Rhesus macaques. The computational strategy consists of two stages: first, MCMC is implemented based on simplified sampling steps, and then, the resulting approximate output is used to generate a proposal distribution for the parameters that results in an efficient ABC procedure. The incorporation of ABC as a correction tool improves the model fit, as is demonstrated through predictive posterior analysis on the data sets of the study.</p><p> Chapter 6 presents additional discussion and comments on potential future research directions.</p> / Dissertation

<p>Analyzing the ever-increasing data of unprecedented scale, dimensionality, diversity, and complexity poses considerable challenges to conventional approaches of statistical modeling. Bayesian nonparametrics constitute a promising research direction, in that such techniques can fit the data with a model that can grow with complexity to match the data. In this dissertation we consider nonparametric Bayesian modeling with completely random measures, a family of pure-jump stochastic processes with nonnegative increments. In particular, we study dictionary learning for sparse image representation using the beta process and the dependent hierarchical beta process, and we present the negative binomial process, a novel nonparametric Bayesian prior that unites the seemingly disjoint problems of count and mixture modeling. We show a wide variety of successful applications of our nonparametric Bayesian latent variable models to real problems in science and engineering, including count modeling, text analysis, image processing, compressive sensing, and computer vision.</p> / Dissertation

Microarray technology has become a dynamic tool in gene expression analysis because it allows for the simultaneous measurement of thousands of gene expressions. Uniqueness in experimental units and microarray data platforms, coupled with how gene expressions are obtained, make the field open for interesting research questions. In this dissertation, we present our investigations of two independent studies related to microarray data analysis. First, we study a recent platform in biology and bioinformatics that compares the quality of genetic information from exfoliated colonocytes in fecal matter with genetic material from mucosa cells within the colon. Using the intraclass correlation coe�cient (ICC) as a measure of reproducibility, we assess the reliability of density estimation obtained from preliminary analysis of fecal and mucosa data sets. Numerical findings clearly show that the distribution is comprised of two components. For measurements between 0 and 1, it is natural to assume that the data points are from a beta-mixture distribution. We explore whether ICC values should be modeled with a beta mixture or transformed first and fit with a normal mixture. We find that the use of mixture of normals in the inverse-probit transformed scale is less sensitive toward model mis-specification; otherwise a biased conclusion could be reached. By using the normal mixture approach to compare the ICC distributions of fecal and mucosa samples, we observe the quality of reproducible genes in fecal array data to be comparable with that in mucosa arrays. For microarray data, within-gene variance estimation is often challenging due to the high frequency of low replication studies. Several methodologies have been developed to strengthen variance terms by borrowing information across genes. However, even with such accommodations, variance may be initiated by the presence of outliers. For our second study, we propose a robust modification of optimal shrinkage variance estimation to improve outlier detection. In order to increase power, we suggest grouping standardized data so that information shared across genes is similar in distribution. Simulation studies and analysis of real colon cancer microarray data reveal that our methodology provides a technique which is insensitive to outliers, free of distributional assumptions, effective for small sample size, and data adaptive.

<p>In the last few decades, data analysis techniques have had to expand to handle large sets of data with complicated structure. This includes identifying low dimensional structure in high dimensional data, analyzing shape and image data, and learning from or classifying large corpora of text documents. Common Bayesian and Machine Learning techniques rely on using the unique geometry of these data types, however departing from Euclidean geometry can result in both theoretical and practical complications. Bayesian nonparametric approaches can be particularly challenging in these areas. </p><p> </p><p>This dissertation proposes a novel approach to these challenges by working with convenient embeddings of the manifold valued parameters of interest, commonly making use of an extrinsic distance or measure on the manifold. Carefully selected extrinsic distances are shown to reduce the computational cost and to increase accuracy of inference. The embeddings are also used to yield straight forward derivations for nonparametric techniques. The methods developed are applied to subspace learning in dimension reduction problems, planar shapes, shape constrained regression, and text analysis.</p> / Dissertation

The purpose of this study was to better understand the goal-striving process in the context of non-problem gambler’s responsible gambling. More specifically, the primary aim of this study was to elucidate the hierarchical structure of goals, the role of the motivational phase of the goal-striving process, and the influence of cognitive evaluation and affective regulation on the goal-striving process. In the first part of the study, a conceptual model is proposed, in which the intrinsic factors used to predict non-problem gamblers’ intentions to gamble responsibly are delineated and tested according to the extension of the theory of planned behavior (Ajzen, 1985, 1991), the model of action phases (Gollwitzer, 1990, 1993), the model of goal-directed behavior (Perugini & Bagozzi, 2001; Perugini & Conner, 2000), and the model of effortful decision making and enactment (Bagozzi, Dholakia, & Basuroy, 2003; Dholakia, Bagozzi, & Gopinath, 2007). Four cognitive factors explain the motivational phase of the goal-striving process, and were incorporated in the current study. One factor explains the goal-oriented behavior at abstract level (i.e., goal feasibility), and the other three explain implementation of action-oriented behavior at concrete level (i.e., attitude toward implementing the actions necessary to achieve the goal, subjective norm, and perceived behavioral control). In addition, two ways of emotional regulation were incorporated to explain the goal-oriented behavior at abstract level. That is, prefactual emotional valence factors related to the success and failure of future goal attainment (anticipated positive and negative emotions) affect goal desire. To sum up, this study anticipated that the proposed antecedent constructs (two anticipated emotions, goal feasibility, attitude, subjective norm, and perceived behavioral control) were strong indicators of how non-problem casino patrons would strive to achieve the goal (i.e., maturing or developing responsible gambling behavior) through a goal-striving process, where the motivational phase plays a critical role in explaining intention to gamble responsibly. A secondary goal of the study was to explore how responsible gambling strategies implemented by the gambling industry influence non-problem casino customers’ goal-directed behavior in a responsible gambling setting. Given the ongoing controversy about the effectiveness of responsible gambling strategies, the focus in the second part of the current study was on how situational arousal factors (i.e., psychological reactance) with regard to external interventions (i.e., compulsory and supplementary responsible gambling strategies) would affect implementation intention, based on the psychological reactance theory (J. W. Brehm, 1989; S. S. Brehm & Brehm, 1981). In other words, situational arousal factors were incorporated herein to explain the extrinsic part of the goal-striving process model. This study was designed to facilitate an understanding of how and why external interventions may fail to deliver the intended effect in the responsible gambling context. In order to take into account the varying effectiveness of responsible gambling strategies, an effort was made to discern between the different effects of each responsible gambling strategy type and to understand in greater detail how these effects were moderated by individual disposition, and especially the strength of the individual’s desire for control. A clear understanding of the moderating effect enables a richer understanding of the effectiveness of responsible gambling strategies with regard to responsible gambling behavior by non-problem casino patrons. Insight gained from the study through analysis of the results is discussed, and important theoretical and practical implications and future research agendas presented in the conclusion. / Tourism and Sport

This study investigated and distinguished between two different aspects of trust (i.e. inter-personal and inter-organisationl) in business relationships. Reviewing the extant literature, a model of business relationships was developed, bridging social exchange theory and transaction cost economics, in addition to using some ideas from the resource-based view. This model was built on an overall framework consisting of three main groups of business relationship characteristics, (1) attitudes (inter-personal and inter-organisational trust) (2) behaviours (commitment, information sharing and relationship-specific investments) and (3) outcomes (financial and non-economic (soft) performance). The overall framework suggested that the attitudinal characteristics affect behavioural characteristics, which consequently affect relationship outcomes. Furthermore, the role of the other party’s opportunistic behaviour as an antecedent of trust aspects was suggested in the model. In line with the overall framework, a basic model was developed with sixteen hypotheses. The model was extended considering dependence as a moderator, and suggesting two additional hypotheses.A questionnaire was designed to measure the characteristics in the model. Data collected from 331 informants (i.e. middle or senior managers knowledgeable about supplier relationships) was used to empirically test the model, using structural equation modeling. The analysis was performed testing the model fit and its underlying hypotheses, additionally using a control variable (the relationship length) and multiple-group analysis (controlling for the size of the company). Inter-personal trust and inter-organisational trust were found to be two distinct constructs (by means of implementing several techniques testing discriminant validity). The results supported the impact of inter-personal trust on inter-organisational trust, inter-personal trust impacting on commitment and information sharing while inter-organisational trust impacting on commitment, information sharing and relationship-specific investments (all in a positive way). The positive effect of behavioural characteristics on relationship outcomes was supported, commitment and relationship-specific investment influencing both financial and non-economic performance, while information sharing influencing non-economic performance. Relationship-specific investments impact positively on commitment, and financial performance impacts positively on non-economic performance. The moderating effects were supported; the positive effect of inter-organisational trust on relationship-specific investments and the positive effect of relationship-specific investments on commitment decrease as dependence increases. The negative effect of the other party’s opportunistic behaviour on trust dimensions was found, while its expected negative effect on relationship-specific investments was not supported in the whole sample. A mixture modeling approach was performed to explore this result. The negative effect was supported in a number of responses, as expected. However, surprisingly a positive effect was also found in a group of responses that were characterised by having relatively longer relationships with the supplier and observing a higher degree of opportunistic behaviour. Additionally relationship-specific investments had a stronger effect on its consequences within this group. Potential explanations for the findings with regard to this group were introduced. The research contributions and implications were also discussed.