Michel Meulders

A Hierarchical IRT Model for Criterion-Referenced Measurement

A hierarchical IRT model is proposed for mastery classification in criterion-referenced measurement. In this model, items measuring the same criterion are grouped, and a difficulty and discrimination parameter of the criterion is estimated on the same scale as the person and item parameters. The level of proficiency of a student with respect to the criterion is determined by the probability of success on the criterion. Cutoff points on the probability scale can be used to classify respondents into masters and nonmasters. The hierarchical IRT model is estimated using the Gibbs sampler and tested using posterior predictive checks. The model is illustrated with a test measuring the attainment targets of reading comprehension (in Dutch) at the end of primary education.

Cross-classification multilevel logistic models in psychometrics.

In IRT models, responses are explained on the basis of person and item effects. Person effects are usually defined as a random sample from a population distribution. Regular IRT models therefore can be formulated as multilevel models, including a within-person part and a between-person part. In a similar way, the effects of the items can be studied as random parameters, yielding multilevel models with a within-item part and a between-item part. The combination of a multilevel model with random person effects and one with random item effects leads to a cross-classification multilevel model, which can be of interest for IRT applications. The use of cross-classification multilevel logistic models will be illustrated with an educational measurement application.

Visual properties of neurons in pulvinar, nucleus lateralis posterior and nucleus suprageniculatus thalami in the cat. I. Qualitative investigation

Summary In ‘cerveau isole’ and pretigeminal cats, about one-third of the neurons located in pulvinar, lateralis posterior, suprageniculatus and posterior nuclei have been shown to react to stationary and/or moving visual stimuli. Some neurons appeared diffusely sensitive to light stimuli, whereas others exhibited restricted visual receptive fields, the mean size of which ranged from arcs of 10°–10°; these visual receptive fields were found most often in the hemifield of vision contralateral to the explored thalamic region. A certain proportion of these visual neurons were maximally excited by stationary stimuli; in many cases they were also discharged by moving patterns but without directional selectivity. The other neurons, however, were strongly activated by stimuli move in a specific direction and with precise orientation with regard to the axis of the displacement. No evident correlation was observed between receptive field position and anatomical localization of the cell; in addition no significant differences could be detected in the visual organization of the different nuclei investigated. Moreover some visually sensitive neurons of the pulvinar were antidromically activated by electrical pulses applied to the suprasylvian gyrus.

Every Cloud Has a Silver Lining: Interpersonal and Individual Differences Determinants of Anger-Related Behaviors:

Two studies examined the effect of status and liking of the anger target on anger behavior and individual differences in angerrelated behavior. Participants recalled anger instances in which the anger target was of higher/equal/lower status and/or liked/ unfamiliar/disliked; subsequently, they indicated which behaviors they had displayed. In both studies, anger behaviors could be grouped into behaviors that imply approaching the target (anger-out, assertion, reconciliation) and behaviors that reflect avoidance/anger-in or social sharing. The results demonstrated that approach behaviors more likely occur toward lower status or liked targets; avoidance behaviors and social sharing more likely occur when the target is of higher status or disliked. On an individual differences level, an approach and an avoid/social sharing person class were identified. The findings suggest that anger may motivate prosocial behavior or social sharing, depending on the individual and type of relation with the target. Only few gender differences were found.

Estimation and software

The aim of this last chapter is threefold. First, we want to give the reader further insights into the estimation methods for the models presented in this volume. Second, we want to discuss the available software for the models presented in this volume. We will not sketch all possibilities of the software, but only those directly relevant to item response modeling as seen in this volume. Third, we want to illustrate the use of various programs for the estimation of a basic model, the Rasch model, for the verbal aggression data.

Eye movement evoked by stimulation of lateral posterior nucleus and pulvinar in the alert cat.

Since Godlowskis studies in 19387, it is known that electrical stimulation of lateral posterior (LP) and pulvinar (Pul.) nuclei in the cat evokes conjugate eye and head movements. These findings were later supported by Roberts 1° who evoked visual searching and stereotyped motor responses in these regions. The participation of the LP-Pul. complex in oculomotor function is not surprising in view of its reciprocal connections with the superior colliculus (SC)1,2, 8. The aim of this study was to describe the eye movements evoked by stimulating the LP-Pul. complex in the alert cat. Experiments were carried out on 3 unanesthetized, restrained adult cats. Eye movements were recorded by 4 previously implanted electro-oculographic (EOG) electrodes. Monopolar electrical stimulation was delivered through tungsten microelectrodes. Negative rectangular pulses of 0.5 msec width at 400 Hz were used in trains of 50 msec to several seconds duration. Current was constantly monitored and never exceeded 1 mA. Restraining methods, EOG calibration and histological procedures have been described previously 11. The animals were maintained in a rather constant state of alertness by a subcutaneous injection of 0.5 mg/kg amphetamine, except in some stimulation sessions in which the influence of variations of alertness was studied. Stimulation of the LP-Pul. complex evoked conjugate, contraversive eye movements of the saccadic type. Except in some cases, which will be detailed below, the velocity of these evoked movements was markedly higher than that of the spontaneous saccades, attaining a velocity of 550 deg/sec for an amplitude of 20 °. Spontaneous saccades of this amplitude do not exceed a speed of 200 deg/sec. Above threshold, saccades appeared as all-or-nothing phenomena. The threshold values ranged, on an average, between 50 and 300 #A. It was however difficult to determine them accurately since each stimulus (though always below 1 mA) did not consistently evoke a saccade. Thresholds were thus chosen as the intensity values which were effective in fairly consistently evoking saccades. No differences in threshold values were observed within the explored nuclei nor did these values vary systematically as a function of electrode depth.

Positive Affect Protects Against Deficient Safety Learning During Extinction of Fear of Movement-Related Pain in Healthy Individuals Scoring Relatively High on Trait Anxiety

UNLABELLED From a treatment perspective, it is highly relevant to pinpoint individual vulnerability factors for resistance to exposure treatment in highly fearful chronic pain patients. Previous fear conditioning research showed that healthy individuals scoring relatively high on trait anxiety display sustained fear to safety cues during extinction. In the context of fear of movement-related pain, this intriguing question has been largely neglected so far. Even more importantly, positive psychological traits such as trait positive affect may function as protective factors against the spreading of fear to safe movements and improve exposure treatment outcomes. In this study, healthy participants completed a trait anxiety and trait positive affect questionnaire and underwent acquisition and extinction of fear of movement-related pain using an experimental voluntary movement paradigm. During acquisition, one movement (CS+) was paired with a painful stimulus and another movement was not (CS-). During extinction, the CS+ was no longer reinforced. Results show failure of fear inhibition to the CS- during extinction in healthy individuals scoring relatively high on trait anxiety or relatively low on positive affect. These findings seem to suggest that safety learning is more vulnerable in healthy people with a high anxious disposition and/or relatively lower levels of positive affect. In addition, this is the first study to show that the negative impact of high trait anxiety on fear inhibition to safety cues during extinction can be countered by high levels of positive affect. These findings may have important clinical implications. PERSPECTIVE Both low positive affect and high trait anxiety are associated with impaired fear inhibition to nonpainful movements during fear extinction. Interestingly, high levels of positive affect buffer against the negative impact of trait anxiety. Increasing positive affect during exposure may counter the effects of trait vulnerabilities and improve treatment outcomes.

Constrained Latent Class Analysis of Three-Way Three-Mode Data

The latent class model for two-way two-mode data can easily be extended to the case of three-way three-mode data as a tool to cluster the elements of one mode on the basis of two other modes simultaneously. However, as the number of manifest variables is typically very large in this type of analysis, the number of conditional probabilities rapidly increases with the number of latent classes, which may lead to an overparameter-ized model. To solve this problem, we introduce a class of constrained latent class models in which the conditional probabilities are a nonlinear function of basic parameters that pertain to each of the two modes. The models can be regarded as a probabilistic extension of related deterministic models or as a generalization of related probabilistic models. For parameter estimation, an EM algorithm can be used to locate the posterior mode, and a Gibbs sampling algorithm can be used to compute a sample of the posterior distribution. Furthermore, model selection criteria and measures to check the fit of the model are discussed. Finally the models are applied to study the types of reactions that occur when one is angry at a person in a certain situation.

LOCALLY DEPENDENT LATENT TRAIT MODEL FOR POLYTOMOUS RESPONSES WITH APPLICATION TO INVENTORY OF HOSTILITY

Psychological tests often involve item clusters that are designed to solicit responses to behavioral stimuli. The dependency between individual responses within clusters beyond that which can be explained by the underlying trait sometimes reveals structures that are of substantive interest. The paper describes two general classes of models for this type of locally dependent responses. Specifically, the models include a generalized log-linear representation and a hybrid parameterization model for polytomous data. A compact matrix notation designed to succinctly represent the system of complex multivariate polytomous responses is presented. The matrix representation creates the necessary formulation for the locally dependent kernel for polytomous item responses. Using polytomous data from an inventory of hostility, we provide illustrations as to how the locally dependent models can be used in psychological measurement.

A Taxonomy of Latent Structure Assumptions for Probability Matrix Decomposition Models.

A taxonomy of latent structure assumptions (LSAs) for probability matrix decomposition (PMD) models is proposed which includes the original PMD model (Maris, De Boeck, & Van Mechelen, 1996) as well as a three-way extension of the multiple classification latent class model (Maris, 1999). It is shown that PMD models involving different LSAs are actually restricted latent class models with latent variables that depend on some external variables. For parameter estimation a combined approach is proposed that uses both a mode-finding algorithm (EM) and a sampling-based approach (Gibbs sampling). A simulation study is conducted to investigate the extent to which information criteria, specific model checks, and checks for global goodness of fit may help to specify the basic assumptions of the different PMD models. Finally, an application is described with models involving different latent structure assumptions for data on hostile behavior in frustrating situations.