Detecting Multiple Random Changepoints in Bayesian Piecewise Growth Mixture Models

Piecewise growth mixture models are a flexible and useful class of methods for analyzing segmented trends in individual growth trajectory over time, where the individuals come from a mixture of two or more latent classes. These models allow each segment of the overall developmental process within each class to have a different functional form; examples include two linear phases of growth, or a quadratic phase followed by a linear phase. The changepoint (knot) is the time of transition from one developmental phase (segment) to another. Inferring the location of the changepoint(s) is often of practical interest, along with inference for other model parameters. A random changepoint allows for individual differences in the transition time within each class. The primary objectives of our study are as follows: (1) to develop a PGMM using a Bayesian inference approach that allows the estimation of multiple random changepoints within each class; (2) to develop a procedure to empirically detect the number of random changepoints within each class; and (3) to empirically investigate the bias and precision of the estimation of the model parameters, including the random changepoints, via a simulation study. We have developed the user-friendly package BayesianPGMM for R to facilitate the adoption of this methodology in practice, which is available at https://github.com/lockEF/BayesianPGMM. We describe an application to mouse-tracking data for a visual recognition task.     

Direction dependence analysis: A framework to test the direction of effects in linear models with an implementation in SPSS

In nonexperimental data, at least three possible explanations exist for the association of two variables x and y: (1) x is the cause of y, (2) y is the cause of x, or (3) an unmeasured confounder is present. Statistical tests that identify which of the three explanatory models fits best would be a useful adjunct to the use of theory alone. The present article introduces one such statistical method, direction dependence analysis (DDA), which assesses the relative plausibility of the three explanatory models on the basis of higher-moment information about the variables (i.e., skewness and kurtosis). DDA involves the evaluation of three properties of the data: (1) the observed distributions of the variables, (2) the residual distributions of the competing models, and (3) the independence properties of the predictors and residuals of the competing models. When the observed variables are nonnormally distributed, we show that DDA components can be used to uniquely identify each explanatory model. Statistical inference methods for model selection are presented, and macros to implement DDA in SPSS are provided. An empirical example is given to illustrate the approach. Conceptual and empirical considerations are discussed for best-practice applications in psychological data, and sample size recommendations based on previous simulation studies are provided.     

TISK 1.0: An easy-to-use Python implementation of the time-invariant string kernel model of spoken word recognition

This article describes a new Python distribution of TISK, the time-invariant string kernel model of spoken word recognition (Hannagan et al. in Frontiers in Psychology, 4, 563, 2013). TISK is an interactive-activation model similar to the TRACE model (McClelland & Elman in Cognitive Psychology, 18, 1–86, 1986), but TISK replaces most of TRACE’s reduplicated, time-specific nodes with theoretically motivated time-invariant, open-diphone nodes. We discuss the utility of computational models as theory development tools, the relative merits of TISK as compared to other models, and the ways in which researchers might use this implementation to guide their own research and theory development. We describe a TISK model that includes features that facilitate in-line graphing of simulation results, integration with standard Python data formats, and graph and data export. The distribution can be downloaded from https://github.com/maglab-uconn/TISK1.0.     

TreeBUGS: An R package for hierarchical multinomial-processing-tree modeling

Multinomial processing tree (MPT) models are a class of measurement models that account for categorical data by assuming a finite number of underlying cognitive processes. Traditionally, data are aggregated across participants and analyzed under the assumption of independently and identically distributed observations. Hierarchical Bayesian extensions of MPT models explicitly account for participant heterogeneity by assuming that the individual parameters follow a continuous hierarchical distribution. We provide an accessible introduction to hierarchical MPT modeling and present the user-friendly and comprehensive R package TreeBUGS, which implements the two most important hierarchical MPT approaches for participant heterogeneity—the beta-MPT approach (Smith & Batchelder, Journal of Mathematical Psychology 54:167-183, 2010) and the latent-trait MPT approach (Klauer, Psychometrika 75:70-98, 2010). TreeBUGS reads standard MPT model files and obtains Markov-chain Monte Carlo samples that approximate the posterior distribution. The functionality and output are tailored to the specific needs of MPT modelers and provide tests for the homogeneity of items and participants, individual and group parameter estimates, fit statistics, and within- and between-subjects comparisons, as well as goodness-of-fit and summary plots. We also propose and implement novel statistical extensions to include continuous and discrete predictors (as either fixed or random effects) in the latent-trait MPT model.     

Brain activation upon ideal-body media exposure and peer feedback in late adolescent girls

Media’s prevailing thin-body ideal plays a vital role in adolescent girls’ body image development, but the co-occurring impact of peer feedback is understudied. The present study used functional magnetic resonance imaging (fMRI) to test media imagery and peer feedback combinations on neural activity related to thin-body ideals. Twenty-four healthy female late adolescents rated precategorized body sizes of bikini models (too thin or normal), directly followed by ostensible peer feedback (too thin or normal). Consistent with prior studies on social feedback processing, results showed increased brain activity in the dorsal medial prefrontal cortex (dmPFC)/anterior cingulate cortex (ACC) and bilateral insula in incongruent situations: when participants rated media models’ body size as normal while peer feedback indicated the models as too thin (or vice versa). This effect was stronger for girls with lower self-esteem. A subsequent behavioral study (N = 34 female late adolescents, separate sample) demonstrated that participants changed behavior in the direction of the peer feedback: precategorized normal sized models were rated as too thin more often after receiving too thin peer feedback. This suggests that the neural responses upon peer feedback may influence subsequent choice. Our results show that media-by-peer interactions have pronounced effects on girls’ body ideals.     

<Literal>flankr</Literal>: An R package implementing computational models of attentional selectivity

The Eriksen flanker task (Eriksen and Eriksen, Perception & Psychophysics, 16, 143–149, 1974) is a classic test in cognitive psychology of visual selective attention. Two recent computational models have formalised the dynamics of the apparent increasing attentional selectivity during stimulus processing, but with very different theoretical underpinnings: The shrinking spotlight (SSP) model (White et al., Cognitive Psychology, 210–238, 2011) assumes attentional selectivity improves in a gradual, continuous manner; the dual stage two phase (DSTP) model (Hübner et al., Psychological Review, 759–784, 2010) assumes attentional selectivity changes from a low- to a high-mode of selectivity at a discrete time-point. This paper presents an R package—flankr—that instantiates both computational models. flankr allows the user to simulate data from both models, and to fit each model to human data. flankr provides statistics of the goodness-of-fit to human data, allowing users to engage in competitive model comparison of the DSTP and the SSP models on their own data. It is hoped that the utility of flankr lies in allowing more researchers to engage in the important issue of the dynamics of attentional selectivity.     

The Factor Structure of the Forms of Self-Criticising/Attacking &amp; Self-Reassuring Scale in Thirteen Distinct Populations

There is considerable evidence that self-criticism plays a major role in the vulnerability to and recovery from psychopathology. Methods to measure this process, and its change over time, are therefore important for research in psychopathology and well-being. This study examined the factor structure of a widely used measure, the Forms of Self-Criticising/Attacking & Self-Reassuring Scale in thirteen nonclinical samples (N?=?7510) from twelve different countries: Australia (N?= 319), Canada (N?= 383), Switzerland (N?= 230), Israel (N?=?476), Italy (N?=?389), Japan (N?=?264), the Netherlands (N?=?360), Portugal (N?=?764), Slovakia (N?=?1326), Taiwan (N?=?417), the United Kingdom 1 (N?=?1570), the United Kingdom 2 (N?=?883), and USA (N?=?331). This study used more advanced analyses than prior reports: a bifactor item-response theory model, a two-tier item-response theory model, and a non-parametric item-response theory (Mokken) scale analysis. Although the original three-factor solution for the FSCRS (distinguishing between Inadequate-Self, Hated-Self, and Reassured-Self) had an acceptable fit, two-tier models, with two general factors (Self-criticism and Self-reassurance) demonstrated the best fit across all samples. This study provides preliminary evidence suggesting that this two-factor structure can be used in a range of nonclinical contexts across countries and cultures. Inadequate-Self and Hated-Self might not by distinct factors in nonclinical samples. Future work may benefit from distinguishing between self-correction versus shame-based self-criticism.     

An automated behavioral measure of mind wandering during computerized reading

Mind wandering is a ubiquitous phenomenon in which attention shifts from task-related to task-unrelated thoughts. The last decade has witnessed an explosion of interest in mind wandering, but research has been stymied by a lack of objective measures, leading to a near-exclusive reliance on self-reports. We addressed this issue by developing an eye-gaze-based, machine-learned model of mind wandering during computerized reading. Data were collected in a study in which 132 participants reported self-caught mind wandering while reading excerpts from a book on a computer screen. A remote Tobii TX300 or T60 eyetracker recorded their gaze during reading. The data were used to train supervised classification models to discriminate between mind wandering and normal reading in a manner that would generalize to new participants. We found that at the point of maximal agreement between the model-based and self-reported mind-wandering means (smallest difference between the group-level means: M _model = .310, M _self = .319), the participant-level mind-wandering proportional distributions were similar and were significantly correlated (r = .400). The model-based estimates were internally consistent (r = .751) and predicted text comprehension more strongly than did self-reported mind wandering (r _model = ?.374, r _self = ?.208). Our results also indicate that a robust strategy of probabilistically predicting mind wandering in cases with poor or missing gaze data led to improved performance on all metrics, as compared to simply discarding these data. Our findings demonstrate that an automated objective measure might be available for laboratory studies of mind wandering during reading, providing an appealing alternative or complement to self-reports.     

Quality of Life in Flemish Families with a Child with an Intellectual Disability: a Multilevel Study on Opinions of Family Members and the Impact of Family Member and Family Characteristics

This study examined family members’ opinions regarding their family quality of life (FQOL). Parents (n = 97) and youngsters (n = 24) of 63 families with a child with an intellectual disability (ID) receiving home-based support completed the Beach Center FQOL Scale. Multilevel models detected substantial variability in FQOL scores and in most subscale scores both within and between families. The observed differences were partly explained by the respondent’s age, the parents’ employment situation, and the support needs of the child with an ID. Role of the family member and gender of the parent could not explain this variability. By including and comparing multiple family members’ opinions, this study contributes to the methodological and conceptual discussions about measuring FQOL.     

A simple computational algorithm of model-based choice preference

A broadly used computational framework posits that two learning systems operate in parallel during the learning of choice preferences—namely, the model-free and model-based reinforcement-learning systems. In this study, we examined another possibility, through which model-free learning is the basic system and model-based information is its modulator. Accordingly, we proposed several modified versions of a temporal-difference learning model to explain the choice-learning process. Using the two-stage decision task developed by Daw, Gershman, Seymour, Dayan, and Dolan (2011), we compared their original computational model, which assumes a parallel learning process, and our proposed models, which assume a sequential learning process. Choice data from 23 participants showed a better fit with the proposed models. More specifically, the proposed eligibility adjustment model, which assumes that the environmental model can weight the degree of the eligibility trace, can explain choices better under both model-free and model-based controls and has a simpler computational algorithm than the original model. In addition, the forgetting learning model and its variation, which assume changes in the values of unchosen actions, substantially improved the fits to the data. Overall, we show that a hybrid computational model best fits the data. The parameters used in this model succeed in capturing individual tendencies with respect to both model use in learning and exploration behavior. This computational model provides novel insights into learning with interacting model-free and model-based components.     

Childhood and Adulthood Socioeconomic Statuses in Relation to Union Interest: Exploratory Surface Relationships

Using response surface methodology for exploratory data, we examined the linear and nonlinear surface relationships between no mobility and mobility in reference to union interest defined by willingness to join a labor union. Analyses were conducted on survey data collected from U.S. nonunion employees (N?=?1,010) who indicated their socioeconomic status (SES) in childhood and adulthood in reference to social class, education, and English language/literacy (e.g., language/literacy during childhood, language/literacy now). Surface relationships were mostly linear. For example, among status indicators, a linear surface was found for employees whose childhood and adulthood language/literacy were the same (no mobility): Lower status nonmobile employees showed more union interest than higher status nonmobile employees. In contrast, a nonlinear surface was found for employees whose childhood and adulthood language/literacy were not the same (mobility): Upwardly mobile employees showed less union interest than downwardly mobile employees showed more interest. Details of the surface relationships and future modeling of the relationships in terms of mediational processes are discussed.     

A class of k-modes algorithms for extracting knowledge structures from data

One of the most crucial issues in knowledge space theory is the construction of the so-called knowledge structures. In the present paper, a new data-driven procedure for large data sets is described, which overcomes some of the drawbacks of the already existing methods. The procedure, called k-states, is an incremental extension of the k-modes algorithm, which generates a sequence of locally optimal knowledge structures of increasing size, among which a “best” model is selected. The performance of k-states is compared to other two procedures in both a simulation study and an empirical application. In the former, k-states displays a better accuracy in reconstructing knowledge structures; in the latter, the structure extracted by k-states obtained a better fit.     

Transitions of Developmental Trajectories of Depressive Symptoms Between Junior and Senior High School Among Youths in Taiwan: Linkages to Symptoms in Young Adulthood

We investigated the heterogeneous developmental trajectories of depressive symptoms in junior and senior high school, the transitions to different trajectories after entering senior high school, and the linkages to the development of depressive symptoms in early adulthood among Taiwanese adolescents. An eight-wave longitudinal data set was analyzed, including 2687 Taiwanese adolescents (51.2% boys, M age?=?14.3 at first wave). Using a manual three-step latent transition growth mixture model, we found that a three-class solution fit the data for both junior high school (termed high-improving, cumulative, and JS-low-stable) and senior high school period (termed heightening, moderate-stable, and HS-low-stable). The depressive symptoms of most individuals maintained at a low level (i.e., low-stable) from adolescence to early adulthood; however, nearly a quarter of the adolescents reported depressive symptoms that were moderately or highly severe in senior high school and beyond. More than 30% of the participants experienced transitioning into a different developmental trajectory between junior and senior high school. When perceiving a higher level of paternal behavioral control, adolescents categorized in the high-improving class in junior high school would have a higher chance to transition to the moderate-stable class than to HS-low-stable class in senior high school. Adolescent boys and girls did not differ in the probability of transitioning between trajectories across junior and senior high school. However, a clear and consistent pattern of symptoms between late adolescence and early adulthood was not observed. These results help elucidate the heterogeneity and fluidity associated with the development of depressive symptoms between early adolescence and early adulthood in light of school transition among youths in Taiwan.     

Hypothesis Testing of the <Emphasis Type="Italic">Q</Emphasis>-matrix

The recent surge of interests in cognitive assessment has led to the development of cognitive diagnosis models. Central to many such models is a specification of the Q-matrix, which relates items to latent attributes that have natural interpretations. In practice, the Q-matrix is usually constructed subjectively by the test designers. This could lead to misspecification, which could result in lack of fit of the underlying statistical model. To test possible misspecification of the Q-matrix, traditional goodness of fit tests, such as the Chi-square test and the likelihood ratio test, may not be applied straightforwardly due to the large number of possible response patterns. To address this problem, this paper proposes a new statistical method to test the goodness fit of the Q-matrix, by constructing test statistics that measure the consistency between a provisional Q-matrix and the observed data for a general family of cognitive diagnosis models. Limiting distributions of the test statistics are derived under the null hypothesis that can be used for obtaining the test p-values. Simulation studies as well as a real data example are presented to demonstrate the usefulness of the proposed method.     

Mixture Models of Categorization

Many currently popular models of categorization are either strictly parametric (e.g., prototype models, decision bound models) or strictly nonparametric (e.g., exemplar models) (F. G. Ashby & L. A. Alfonso-Reese, 1995, Journal of Mathematical Psychology, 39, 216-233). In this article, a family of semiparametric classifiers is investigated where categories are represented by a finite mixture distribution. The advantage of these mixture models of categorization is that they contain several parametric models and nonparametric models as a special case. Specifically, it is shown that both decision bound models (F. G. Ashby & W. T. Maddox, 1992, Journal of Experimental Psychology: Human Perception and Performance, 16, 598-612; 1993, Journal of Mathematical Psychology, 37, 372-400) and the generalized context model (R. M. Nosofsky, 1986, Journal of Experimental Psychology: General, 115, 39-57) can be interpreted as two extreme cases of a common mixture model. Furthermore, many other (semiparametric) models of categorization can be derived from the same generic mixture framework. In this article, several examples are discussed and a parameter estimation procedure for fitting these models is outlined. To illustrate the approach, several specific models are fitted to a data set collected by S. C. McKinley and R. M. Nosofsky (1995, Journal of Experimental Psychology: Human Perception and Performance, 21, 128-148). The results suggest that semi-parametric models are a promising alternative for future model development.     

Estimating the average need of semantic knowledge from distributional semantic models

Continuous bag of words (CBOW) and skip-gram are two recently developed models of lexical semantics (Mikolov, Chen, Corrado, & Dean, Advances in Neural Information Processing Systems, 26, 3111–3119, 2013). Each has been demonstrated to perform markedly better at capturing human judgments about semantic relatedness than competing models (e.g., latent semantic analysis; Landauer & Dumais, Psychological Review, 104(2), 1997 211; hyperspace analogue to language; Lund & Burgess, Behavior Research Methods, Instruments, & Computers, 28(2), 203–208, 1996). The new models were largely developed to address practical problems of meaning representation in natural language processing. Consequently, very little attention has been paid to the psychological implications of the performance of these models. We describe the relationship between the learning algorithms employed by these models and Anderson’s rational theory of memory (J. R. Anderson & Milson, Psychological Review, 96(4), 703, 1989) and argue that CBOW is learning word meanings according to Anderson’s concept of needs probability. We also demonstrate that CBOW can account for nearly all of the variation in lexical access measures typically attributable to word frequency and contextual diversity—two measures that are conceptually related to needs probability. These results suggest two conclusions: One, CBOW is a psychologically plausible model of lexical semantics. Two, word frequency and contextual diversity do not capture learning effects but rather memory retrieval effects.     

Eight-Year Latent Class Trajectories of Academic and Social Functioning in Children with Attention-Deficit/Hyperactivity Disorder

We examined trajectories of academic and social functioning in children with attention-deficit/hyperactivity disorder (ADHD) to identify those who might be at risk for especially severe levels of academic and social impairment over time. We estimated a series of growth mixture models using data from two subsamples of children participating in the NIMH Collaborative Multisite Multimodal Treatment Study of Children with ADHD (MTA) including those with at least baseline and 96-month data for reading and mathematics achievement (n = 392; 77.3% male; M age = 7.7; SD = 0.8) or social skills ratings from teachers (n = 259; 74.9% male; M age = 7.6; SD = 0.8). We compared latent trajectories for children with ADHD to mean observed trajectories obtained from a local normative (i.e., non-ADHD) comparison group (n = 289; 80.6% male; M age = 9.9; SD = 1.1). Results indicated six latent trajectory classes for reading and mathematics and four classes for teacher social skills ratings. There was not only a relationship between trajectories of inattention symptoms and academic impairment, but also a similarly strong association between trajectory classes of hyperactive-impulsive symptoms and achievement. Trajectory class membership correlated with socio-demographic and diagnostic characteristics, inattention and hyperactive-impulsive symptom trajectories, externalizing behavior in school, and treatment receipt and dosage. Although children with ADHD display substantial heterogeneity in their reading, math, and social skills growth trajectories, those with behavioral and socio-demographic disadvantages are especially likely to display severe levels of academic and social impairment over time. Evidence-based early screening and intervention that directly address academic and social impairments in elementary school-aged children with ADHD are warranted. The ClinicalTrials.gov identifier is NCT00000388.     

Model comparison in ANOVA

Analysis of variance (ANOVA), the workhorse analysis of experimental designs, consists of F-tests of main effects and interactions. Yet, testing, including traditional ANOVA, has been recently critiqued on a number of theoretical and practical grounds. In light of these critiques, model comparison and model selection serve as an attractive alternative. Model comparison differs from testing in that one can support a null or nested model vis-a-vis a more general alternative by penalizing more flexible models. We argue this ability to support more simple models allows for more nuanced theoretical conclusions than provided by traditional ANOVA F-tests. We provide a model comparison strategy and show how ANOVA models may be reparameterized to better address substantive questions in data analysis.     

Executive Functioning Correlates of <Emphasis Type="Italic">DSM-5</Emphasis> Maladaptive Personality Traits: Initial Evidence from an Italian Sample of Consecutively Admitted Adult Outpatients

In order to evaluate the associations between computer-administered tasks of executive functioning (EF), and maladaptive personality domains and traits listed in DSM-5 Alternative Model of Personality Disorders, 53 consecutively admitted psychotherapy outpatients (female participants: n?=?27, 50.9%; male participants: n?=?26, 49.1%; participants’ mean age?=?37.28 years, SD?=?11.50 years) were administered the Psychology Experiment Building Language (PEBL) EF tasks and the Personality Inventory for DSM-5 (PID-5). According to rank-order correlation analyses, a number of non-negligible and specific associations were observed between selected PID-5 scales and indices of participants’ performance on EF tasks. MM robust regression models showed that participants’ performance on computer-administered EF tasks explained a non-negligible amount of variance in selected PID-5 scale scores (median R² value?=?.17). As a whole, our trait-level analyses of PID-5 dimensions suggest the clinical usefulness of integrating self-reports and EF laboratory tasks in routine clinical assessment.