Showing papers by "Paul De Boeck published in 1999"

Indclas: a three-way hierarchical classes model

Abstract: A three-way three-mode extension of De Boeck and Rosenberg's (1988) two-way two-mode hierarchical classes model is presented for the analysis of individual differences in binary object × attribute arrays. In line with the two-way hierarchical classes model, the three-way extension represents both the association relation among the three modes and the set-theoretical relations among the elements of each model. An algorithm for fitting the model is presented and evaluated in a simulation study. The model is illustrated with data on psychiatric diagnosis. Finally, the relation between the model and extant models for three-way data is discussed.

Confirmatory Analyses of Componential Test Structure Using Multidimensional Item Response Theory.

Abstract: The componential structure of synonym tasks is investigated using confirmatory multidimensional two-parameter IRT models. It was hypothesized that an open synonym task is decomposable into generating synonym candidates and evaluating these candidate words with respect to their synonymy with the stimulus word. Two subtasks were constructed to identify these two components. Different confirmatory models were estimated both with TESTMAP and with NOHARM. The componential hypothesis was supported, but it was found that the generation subtask also involved some evaluation and that generation and evaluation were highly correlated.

Predicting conjunction typicalities by component typicalities.

Abstract: In two studies, we investigated to what extent typicalities in conjunctive concepts phrased as relative clauses—such aspets that are also birds—can be predicted from simple functions of constituent typicalities and from extensions of such functions. In a first study, analyses of a large aggregated data set, based on seven different experiments, showed that a calibrated minimum rule model and some extensions of this model accounted for a very large part of the variance in the conjunction typicalities. The same models can also account for the so-called guppy effect. A psychological explanation is presented, which states that typicalities in contrast categories, likepets that are not birds andbirds that are not pets, further improve the prediction of conjunction typicalities. This hypothesis is tested in a second study.

A two-process theory of facial perception of emotions.

Abstract: Atwo-process probabilistic theory of emotion perception based on a non-linear combination of facial features is presented. Assuming that the upper and the lower part of the face function as the building blocks at the basis of emotion perception, an empirical test is provided with fear and happiness as target emotions. Subjects were presented with prototypical fearful and happy faces and with computer-generated chimerical expressions that were a combination of happy and fearful. Subjects were asked to indicate the emotions they perceive using an extensive list of emotions. We show that some emotions require a conjunction of the two halves of a face to be perceived, whereas for some other emotions only one half is sufficient. We demonstrate that chimerical faces give rise to the perception of genuine emotions. The findings provide evidence that different combinations of the two halves of a fearful and a happy face, either congruent or not, do generate the perception of emotions other than fear and happiness.

A Generic DisjunctiveConjunctive Decomposition Model for n-ary Relations

Abstract: This paper discusses a generic decomposition model that represents an arbitrary n-ary relation as a disjunctive or conjunctive combination of a number of n-ary component relations of a prespecified type. An important subclass of order-preserving decompositions is defined and its properties are derived. The generic model is shown to subsume various known models as special cases, including the models of Boolean factor analysis, hierarchical classes analysis, and disjunctiveconjunctive nonmetric factor analysis. Moreover, it also subsumes a broad range of new models as exemplified with a novel model for multidimensional parallelogram analysis and novel threeway extensions of nonmetric factor analysis. 1999 Academic Press

Distinguishing Constant and Dimension-Dependent Interaction: A Simulation Study.

Abstract: A simulation study was conducted to determine how well two models for local item dependency (LID), called interaction models, could be distinguished. The models examined were the constantorder interaction model (COIM) and the dimension dependent interaction model (DDIM). Data were simulated according to the latter model. Three factors were manipulated: sample size, the weight of the difference between the latent trait value of the examinee and the interaction parameter, and the value of the interaction parameter. Results indicated that (1) if the interaction parameter is not too extreme, the COIM will be rejected in favor of the true model (the Rasch model fit poorly for all levels of the interaction parameter); (2) a larger weight of the difference between the latent trait value and the interaction parameter facilitated the rejection of the COIM, although finding the true weight required a large sample size; and (3) the value for the interaction parameter with an optimal discrimination between the COIM a...