Stochastic ordering using the latent trait and the sum score in polytomous IRT models

In a restricted class of item response theory (IRT) models for polytomous items the unweighted total score has monotone likelihood ratio (MLR) in the latent trait. MLR implies two stochastic ordering (SO) properties, denoted SOM and SOL, which are both weaker than MLR, but very useful for measurement with IRT models. Therefore, these SO properties are investigated for a broader class of IRT models for which the MLR property does not hold.In this study, first a taxonomy is given for nonparametric and parametric models for polytomous items based on the hierarchical relationship between the models. Next, it is investigated which models have the MLR property and which have the SO properties. It is shown that all models in the taxonomy possess the SOM property. However, counterexamples illustrate that many models do not, in general, possess the even more useful SOL property.Hemker's research was supported by the Netherlands Research Council, Grant 575-67-034. Junker's research was supported in part by the National Institutes of Health, Grant CA54852, and by the National Science Foundation, Grant DMS-94.04438.     

Perceptual differentiation as a source of category effects in object processing: Evidence from naming and object decision

The locus of category effects in picture recognition and naming was examined in two experiments with normal subjects. Subjects carried out object decision (deciding whether the stimulus is a “real” object or not) and naming tasks with pictures of clothing, furniture, fruit, and vegetables. These categories are distinguished by containing either relatively many exemplars with similar perceptual structures (fruit and vegetables;structurally similar categories), or relatively few exemplars with similar perceptual structures (clothing and furniture;structurally dissimilar categories). In Experiment 1, responses to the stimuli from the structurally similar categories were slower than responses to stimuli from the structurally dissimilar categories, and this effect was larger in the naming than in the object decision task. Further, prior object decisions to stimuli from structurally similar categories facilitated their subsequent naming. In Experiment 2, we orthogonally manipulated object decision and naming as prime and target tasks, again with stimuli from the four categories. Category effects, with responses slower to objects from structurally similar categories, were again larger in naming than in object decision, and these category effects in naming were reduced by priming with both naming and object decision. We interpret the data to indicate that category effects in object naming can reflect visually based competition which is reduced by the preactivation of stored structural knowledge for objects.     

Dysphoria and Children's Processing of Supportive Interactions

This study examined the processing of supportive interactions by dysphoric and nondysphoric preteens and early adolescents. Seventy-two youngsters between the ages of 10 and 13 evaluated the supportiveness and helpfulness of standardized, videotaped interactions between a distressed preadolescent and a maternal figure. The tape presentations varied in terms of the level of depicted maternal support and instructional condition (degree of self-reference). The results indicated that dysphoric youngsters evaluated both the supportiveness and helpfulness of interactions less positively than nondysphoric agemates. Group differences in support evaluations were most pronounced in the self-referenced condition. The level of depicted support did not affect processing differences. Dysphoric subjects reported lower levels of emotional support in prior relationships and a greater tendency to view supportive behavior as ingenuine than nondysphoric peers. Variation in prior support experiences accounted for group differences in the evaluation of the supportiveness of new interactions.     

“Are False Memories Permanent?”: An Investigation of the Long-Term Effects of Source Misattributions

With growing concerns over children's suggestibility and how it may impact their reliability as witnesses, there is increasing interest in determining the long-term effects of induced memories. The goal of the present research was to learn whether source misattributions found by Ceci, Huffman, Smith, and Loftus (1994) caused permanent memory alterations in the subjects tested. When 22 children from the original study were reinterviewed 2 years later, they recalled 77% of all true events. However, they only consented to 13% of all false events, compared to the 22% false consent rate (among the same subset of children) found by Ceci et al. (1994). Additionally, while children remained accurate in their recall of true events (they maintained assents 78% of the time), they “recanted” their earlier false consents 77% of the time, after the 2-year delay. Implications of these findings for child witnesses and the legal system are discussed.     

Neural Network Models and Mechanisms of Strategy Development

The focus of this paper is mechanisms that may be responsible for intellectual and developmental differences in the cognitive strategies of typical and atypical children, including those with mental retardation. The discussion of these mechanisms is based on behavioral experiments on external memory strategies and on a set of neural network models designed for these tasks. Following the review of the external memory research, the rationale for using neural network models, how they have been used in other research, and their specific application to intellectual and developmental differences in external memory, including the results of several simulations, are reviewed. This is followed by a discussion of the mechanisms of intellectual differences and developmental change included in the models and some challenges for this type of modeling. Neural network modeling is discussed as an asset to research on cognitive development.