Order‐constrained linear optimization

Despite the fact that data and theories in the social, behavioural, and health sciences are often represented on an ordinal scale, there has been relatively little emphasis on modelling ordinal properties. The most common analytic framework used in psychological science is the general linear model, whose variants include ANOVA, MANOVA, and ordinary linear regression. While these methods are designed to provide the best fit to the metric properties of the data, they are not designed to maximally model ordinal properties. In this paper, we develop an order‐constrained linear least‐squares (OCLO) optimization algorithm that maximizes the linear least‐squares fit to the data conditional on maximizing the ordinal fit based on Kendall's τ. The algorithm builds on the maximum rank correlation estimator (Han, 1987, Journal of Econometrics, 35, 303) and the general monotone model (Dougherty & Thomas, 2012, Psychological Review, 119, 321). Analyses of simulated data indicate that when modelling data that adhere to the assumptions of ordinary least squares, OCLO shows minimal bias, little increase in variance, and almost no loss in out‐of‐sample predictive accuracy. In contrast, under conditions in which data include a small number of extreme scores (fat‐tailed distributions), OCLO shows less bias and variance, and substantially better out‐of‐sample predictive accuracy, even when the outliers are removed. We show that the advantages of OCLO over ordinary least squares in predicting new observations hold across a variety of scenarios in which researchers must decide to retain or eliminate extreme scores when fitting data.     

An improved algorithm for predicting free recalls

Laming [Laming, D. (2006). Predicting free recalls. Journal of Experimental Psychology: Learning, Memory, and Cognition, 32, 1146-1163] has shown that, in a free-recall experiment in which the participants rehearsed out loud, entire sequences of recalls could be predicted, to a useful degree of precision, from the prior sequences of stimuli and rehearsals. This article describes an improved predictive algorithm, which is then used to re-analyse two further sets of free-recall data. Prediction is compared, generically, with conventional modelling, represented here by three recent models of free recall that are concerned with recalls only, not with rehearsals. Some implications are drawn, from the use of rehearsal data by the predictive algorithm, for some of the constituent assumptions embodied in those three models.     

Unsupervised Classification During Time-Series Model Building

Researchers who collect multivariate time-series data across individuals must decide whether to model the dynamic processes at the individual level or at the group level. A recent innovation, group iterative multiple model estimation (GIMME), offers one solution to this dichotomy by identifying group-level time-series models in a data-driven manner while also reliably recovering individual-level patterns of dynamic effects. GIMME is unique in that it does not assume homogeneity in processes across individuals in terms of the patterns or weights of temporal effects. However, it can be difficult to make inferences from the nuances in varied individual-level patterns. The present article introduces an algorithm that arrives at subgroups of individuals that have similar dynamic models. Importantly, the researcher does not need to decide the number of subgroups. The final models contain reliable group-, subgroup-, and individual-level patterns that enable generalizable inferences, subgroups of individuals with shared model features, and individual-level patterns and estimates. We show that integrating community detection into the GIMME algorithm improves upon current standards in two important ways: (1) providing reliable classification and (2) increasing the reliability in the recovery of individual-level effects. We demonstrate this method on functional MRI from a sample of former American football players.     

Symptom schemata in chronic respiratory patients

In view of evidence that illness prognoses and adaptive functioning may be influenced by the accuracy of people's knowledge about their physical symptoms, the present study extended these findings to the chronic care population. It was hypothesized that individuals hold beliefs and develop theories about their physical symptoms and that the accuracy of these beliefs is predictive of the individuals' general level of adaptive functioning. A modified version of an instrument designed to measure the accuracy of illness schemata was employed with a sample of 31 chronic respiratory patients. Accuracy rating correlated positively and significantly with ratings of adaptive functioning, whereas no relationship was observed between adaptive functioning and severity of the patients' medical condition. Well-informed patients functioned at a higher level physically, psychologically, and socially than less-informed patients. These findings point to the importance of patient education for prognosis. Possible mediating factors are discussed.     

Job aspects in the School Psychology Service: Empirically distinct associations with positive challenge at work,perceived control at work,and job attitudes

Activities within the Norwegian School Psychology Service traditionally concentrate on assessment of individuals in need and report writing (individual-level treatment). In recent years, a more “systemic” job aspect, providing preschool/school staff with tools to prevent the emergence of problems (systemic-level prevention) has been in demand. This study investigates the predictive value of these distinct aspects of work behaviour in the School Psychology Service using the outcome variables job satisfaction, job involvement, and organizational commitment among employees. It also looks at to what extent such associations may be mediated by positive challenge at work and perceived control at work. Self-reported data were collected from 470 School Psychology counsellors within 121 offices. Structural equation modelling suggested that the preventive aspect of the job was positively related to job attitudes, while the traditional job aspect was not. Furthermore, these associations were partially mediated by positive challenge at work and perceived control at work. Control for age did not alter the magnitude or pattern of regression coefficients in any substantive way. Occupational group did not moderate the associations. Alternative models were tested. Implications are discussed.     

Predictive coding as a model of cognition

Previous work has shown that predictive coding can provide a detailed explanation of a very wide range of low-level perceptual processes. It is also widely believed that predictive coding can account for high-level, cognitive, abilities. This article provides support for this view by showing that predictive coding can simulate phenomena such as categorisation, the influence of abstract knowledge on perception, recall and reasoning about conceptual knowledge, context-dependent behavioural control, and naive physics. The particular implementation of predictive coding used here (PC/BC-DIM) has previously been used to simulate low-level perceptual behaviour and the neural mechanisms that underlie them. This algorithm thus provides a single framework for modelling both perceptual and cognitive brain function.     

School and individual factors on game addiction: A multilevel analysis

This study investigated the protective or risk factors of game addiction at individual and school-level using multilevel modelling. Data from Korea Children and Youth Panel Survey was used for the analyses. The findings revealed that individual-level variables, including gender, delinquency, family intimacy, stress, self-esteem and self-control, were significant predictors of game addiction. Furthermore, protective school climate contributed to decreasing the level of game addiction and risky school climates positively predicted game addiction. The effect of individual-level variables on game addiction was different depending on the type or level of school-level variables. Based on the findings, implications and suggestions for future studies were discussed.     

Evaluation of latent construct correlations in the presence of missing data: a note on a latent variable modelling approach

A latent variable modelling approach is discussed, which can be used to evaluate indices of linear relationship between latent constructs in incomplete data sets. The method is based on an application of maximum-likelihood estimation and inclusion of covariates predictive of missing values. The approach can be employed for point and interval estimation of latent correlations in the presence of missing data, and capitalizes on enhanced plausibility of the assumption of data missing at random through introduction of informative covariates. The method is illustrated on empirical data.     

Toward a predictive model of suicidal ideation and behavior: some preliminary data in college students

Within a multivariate statistical design, the relationship of various interpersonal, emotional, and cognitive variables to suicidal ideation and behavior in college students was examined. A total of 158 subjects (58 males, 100 females) completed self-report measures of life stress, faulty cognitions, loneliness, depression, hopelessness, family cohesiveness, adaptive reasons for living, and suicidal ideation and behavior. The results of a multiple-regression analysis, forward-inclusion algorithm, indicated that a predictive equation consisting of loneliness, irrational beliefs, and low adaptive reasons for living best accounted for suicidal behavior scores. To determine the basic structures and power of the predictor variables under study, a factor analysis and composite regression were employed, resulting in a predictive equation consisting of three factors: Social/Emotional Alienation, Cognitive Distortions, and Deficient Adaptive Resources. Results are discussed in terms of an interactional model of suicidal behavior.     

Modelling intransitive preferences: A random-effects approach

Current models for individual-level paired comparison data are based on the three assumptions that (1) pairwise judgments are independent, (2) the utility of an item remains invariant across trials, and (3) pair-specific variability can account for intransitive choice behaviour. All three assumptions seem strong and likely to be violated in empirical applications. This paper introduces a new framework for the analysis of paired comparison data which relaxes these three assumptions and considers the utilities associated with the same item across trials to be neither independent nor identical, but related. The proposed approach provides new insights about the reliability and consistency of paired comparison judgments and can account for systematic violations of transitivity. An application, based on a replication of Tversky's [(1969). Intransitivity of preference. Psychological Review, 76, 31-48] gamble study, illustrates the usefulness of the new approach in modelling both transitive and intransitive preferences.     

The predictive accuracy of intertemporal‐choice models

How do people choose between a smaller reward available sooner and a larger reward available later? Past research has evaluated models of intertemporal choice by measuring goodness of fit or identifying which decision‐making anomalies they can accommodate. An alternative criterion for model quality, which is partly antithetical to these standard criteria, is predictive accuracy. We used cross‐validation to examine how well 10 models of intertemporal choice could predict behaviour in a 100‐trial binary‐decision task. Many models achieved the apparent ceiling of 85% accuracy, even with smaller training sets. When noise was added to the training set, however, a simple logistic‐regression model we call the difference model performed particularly well. In many situations, between‐model differences in predictive accuracy may be small, contrary to long‐standing controversy over the modelling question in research on intertemporal choice, but the simplicity and robustness of the difference model recommend it to future use.     

Preference‐based Learning of Ideal Solutions in TOPSIS‐like Decision Models

Combining established modelling techniques from multiple‐criteria decision aiding with recent algorithmic advances in the emerging field of preference learning, we propose a new method that can be seen as an adaptive version of TOPSIS, the technique for order preference by similarity to ideal solution decision model (or at least a simplified variant of this model). On the basis of exemplary preference information in the form of pairwise comparisons between alternatives, our method seeks to induce an ‘ideal solution’ that, in conjunction with a weight factor for each criterion, represents the preferences of the decision maker. To this end, we resort to probabilistic models of discrete choice and make use of maximum likelihood inference. First experimental results on suitable preference data suggest that our approach is not only intuitively appealing and interesting from an interpretation point of view but also competitive to state‐of‐the‐art preference learning methods in terms of prediction accuracy. Copyright © 2014 John Wiley & Sons, Ltd.     

Do complex models increase prediction of complex behaviours? Predicting driving ability in people with brain disorders

Prediction of complex behavioural tasks via relatively simple modelling techniques, such as logistic regression and discriminant analysis, often has limited success. We hypothesized that to more accurately model complex behaviour, more complex models, such as kernel-based methods, would be needed. To test this hypothesis, we assessed the value of six modelling approaches for predicting driving ability based on performance on computerized sensory–motor and cognitive tests (SMCTests?) in 501 people with brain disorders. The models included three models previously used to predict driving ability (discriminant analysis, DA; binary logistic regression, BLR; and nonlinear causal resource analysis, NCRA) and three kernel methods (support vector machine, SVM; product kernel density, PK; and kernel product density, KP). At the classification level, two kernel methods were substantially more accurate at classifying on-road pass or fail (SVM 99.6%, PK 99.8%) than the other models (DA 76%, BLR 78%, NCRA 74%, KP 81%). However, accuracy decreased substantially for all of the kernel models when cross-validation techniques were used to estimate prediction of on-road pass or fail in an independent referral group (SVM 73–76%, PK 72–73%, KP 71–72%) but decreased only slightly for DA (74–75%) and BLR (75–76%). Cross-validation of NCRA was not possible. In conclusion, while kernel-based models are successful at modelling complex data at a classification level, this is likely to be due to overfitting of the data, which does not lead to an improvement in accuracy in independent data over and above the accuracy of other less complex modelling techniques.     

Do complex models increase prediction of complex behaviours? Predicting driving ability in people with brain disorders

Prediction of complex behavioural tasks via relatively simple modelling techniques, such as logistic regression and discriminant analysis, often has limited success. We hypothesized that to more accurately model complex behaviour, more complex models, such as kernel-based methods, would be needed. To test this hypothesis, we assessed the value of six modelling approaches for predicting driving ability based on performance on computerized sensory-motor and cognitive tests (SMCTests?) in 501 people with brain disorders. The models included three models previously used to predict driving ability (discriminant analysis, DA; binary logistic regression, BLR; and nonlinear causal resource analysis, NCRA) and three kernel methods (support vector machine, SVM; product kernel density, PK; and kernel product density, KP). At the classification level, two kernel methods were substantially more accurate at classifying on-road pass or fail (SVM 99.6%, PK 99.8%) than the other models (DA 76%, BLR 78%, NCRA 74%, KP 81%). However, accuracy decreased substantially for all of the kernel models when cross-validation techniques were used to estimate prediction of on-road pass or fail in an independent referral group (SVM 73-76%, PK 72-73%, KP 71-72%) but decreased only slightly for DA (74-75%) and BLR (75-76%). Cross-validation of NCRA was not possible. In conclusion, while kernel-based models are successful at modelling complex data at a classification level, this is likely to be due to overfitting of the data, which does not lead to an improvement in accuracy in independent data over and above the accuracy of other less complex modelling techniques.     

Assessment of generalised Bayesian structural equation models for continuous and binary data

The paper proposes a novel model assessment paradigm aiming to address shortcoming of posterior predictive $p$-values, which provide the default metric of fit for Bayesian structural equation modelling (BSEM). The model framework presented in the paper focuses on the approximate zero approach (Psychological Methods, 17 , 2012, 313), which involves formulating certain parameters (such as factor loadings) to be approximately zero through the use of informative priors, instead of explicitly setting them to zero. The introduced model assessment procedure monitors the out-of-sample predictive performance of the fitted model, and together with a list of guidelines we provide, one can investigate whether the hypothesised model is supported by the data. We incorporate scoring rules and cross-validation to supplement existing model assessment metrics for BSEM. The proposed tools can be applied to models for both continuous and binary data. The modelling of categorical and non-normally distributed continuous data is facilitated with the introduction of an item-individual random effect. We study the performance of the proposed methodology via simulation experiments as well as real data on the ‘Big-5’ personality scale and the Fagerstrom test for nicotine dependence.     

Construct-oriented Biodata: Capturing Change-related and Contextually Relevant Future Performance

This study highlights an approach to the development of performance and predictor constructs that has the potential to permit context and culture relevant selection. Task, contextual and adaptive measures of performance were used as criterion data in a selection study involving 325 staff in the hotel industry. Construct-oriented biodata were developed to predict adaptive and contextual performance and the validities were contrasted with measures of cognitive ability and personality. The results provided construct support for the separation of adaptive performance from task and contextual performance and for the predictive validity of change-related biodata after controlling for cognitive ability. The pattern of correlations among the personality measures, biodata scales and cognitive tests supported the construct validity of the biodata scales. Results are discussed in relation to the context in which the data were collected, and the general relevance of the procedure across countries or cultures.     

High-dimensional maximum marginal likelihood item factor analysis by adaptive quadrature

Although the Bock–Aitkin likelihood-based estimation method for factor analysis of dichotomous item response data has important advantages over classical analysis of item tetrachoric correlations, a serious limitation of the method is its reliance on fixed-point Gauss-Hermite (G-H) quadrature in the solution of the likelihood equations and likelihood-ratio tests. When the number of latent dimensions is large, computational considerations require that the number of quadrature points per dimension be few. But with large numbers of items, the dispersion of the likelihood, given the response pattern, becomes so small that the likelihood cannot be accurately evaluated with the sparse fixed points in the latent space. In this paper, we demonstrate that substantial improvement in accuracy can be obtained by adapting the quadrature points to the location and dispersion of the likelihood surfaces corresponding to each distinct pattern in the data. In particular, we show that adaptive G-H quadrature, combined with mean and covariance adjustments at each iteration of an EM algorithm, produces an accurate fast-converging solution with as few as two points per dimension. Evaluations of this method with simulated data are shown to yield accurate recovery of the generating factor loadings for models of upto eight dimensions. Unlike an earlier application of adaptive Gibbs sampling to this problem by Meng and Schilling, the simulations also confirm the validity of the present method in calculating likelihood-ratio chi-square statistics for determining the number of factors required in the model. Finally, we apply the method to a sample of real data from a test of teacher qualifications.     

Making the future memorable: The phenomenology of remembered future events

Although our ability to remember future simulations conveys an adaptive advantage, enabling us to better prepare for upcoming events, the factors influencing the memorability of future simulations are not clear. In this study, participants generated future simulations that combined specific people, places and objects from memory, and for each trial, made a series of phenomenological ratings about the event components and the simulation as a whole. Memory for simulations was later assessed using a cued-recall test. We used multilevel modelling to determine whether the phenomenological qualities of event components (familiarity, emotionality and significance) and simulations (detail, plausibility) were predictive of whether the simulation was successfully encoded and later accessible. Our results demonstrate that person familiarity, detail and plausibility were significant predictors of whether a given future simulation was encoded into memory and later accessible. These findings suggest that scaffolding future simulations with pre-existing episodic memories is the path to a memorable future.     

Adolescent Family Adversity and Mental Health Problems: The Role of Adaptive Self-regulation Capacities. The TRAILS Study

Adolescent family adversity is a considerable adaptive challenge in an increasingly turbulent developmental period. Using data from a prospective population cohort of 2230 Dutch adolescents, we tested risk-buffering interactions between adolescent family adversity and self-regulation capacities on mental health. We used two adaptive self-regulation capacities that could allow adolescents to manage relatively well with family adversity: (1) parent-reported effortful control, and (2) an attentional flexibility (in this case, set-shifting) task. Adolescent family adversity was associated with internalizing problems and externalizing problems. The risk-buffering effects of effortful control were found for externalizing problems but not for internalizing problems. There were no risk-buffering effects of attentional flexibility on both types of mental health problems. Effortful control is likely to benefit adolescents’ ability to channel their frustrations in adaptive ways in the presence of family adversity. Additionally, (attentional) set-shifting tasks might have a limited predictive value for risk-buffering research.