Simple additivity of stochastic psychological processes: Tests and measures

In this article, some additive models of behavioral measures are defined, and distributional tests of them are proposed. Major theoretical results include (a) conditions for additivity of components to predict the highest level of dominance in a model-free stochastic dominance hierarchy, (b) methods of identifying the shape of the hazard rate function of an added component from certain relationships among the observable density and distribution functions, and (c) effects of stochastic dependence between components on the distributional tests. Feasibility and usefulness of the methods are demonstrated by application to choice RT and ratings experiments.The author was supported by grants MH44640 to Roger Ratcliff and AFOSR 90-0246 (jointly funded by NSF) to Gail McCoon. Parts of this work were presented at the European Mathematical Psychology meetings of 1991. My thanks to F. Gregory Ashby, Bruce Bloxom, Roger Ratcliff, W. Schwarz, Jim Townsend, and an anonymous reviewer for their many helpful suggestions.     

Revisiting “Kindergarten as academic boot camp”: a nationwide study of ability grouping and psycho-social development

We revisit Harry L. Gracey’s perspective of kindergarten as academic boot camp where, at school entry, children acquire the student role through a structured program of activities. We provide further insights into the crucial mechanisms of socialization that occur in U.S. kindergartens by examining the relationship between within-class ability grouping and attributes of children’s psycho-social development that are critical for school success. Using data from the Early Childhood Longitudinal Study (ECLS-K) and incorporating stratified propensity scores into multi-level models, our analyses link within-class ability grouping to student behaviors in a manner that supports the concerns raised by ability grouping critics. Children in low ability groups show slower development in approaches to learning than comparable ungrouped children. Children in high ability groups show more accelerated development in all psycho-social attributes considered (approaches to learning, interpersonal skills, self control and externalizing problem behaviors) than comparable ungrouped children. These asymmetrical findings suggest that children placed in high and low ability groups are socialized into different versions of the student role.     

The Role of Conditional Likelihoods in Latent Variable Modeling

Psychometrika - In psychometrics, the canonical use of conditional likelihoods is for the Rasch model in measurement. Whilst not disputing the utility of conditional likelihoods in measurement, we...     

Additive Multilevel Item Structure Models with Random Residuals: Item Modeling for Explanation and Item Generation

An additive multilevel item structure (AMIS) model with random residuals is proposed. The model includes multilevel latent regressions of item discrimination and item difficulty parameters on covariates at both item and item category levels with random residuals at both levels. The AMIS model is useful for explanation purposes and also for prediction purposes as in an item generation context. The parameters can be estimated with an alternating imputation posterior algorithm that makes use of adaptive quadrature, and the performance of this algorithm is evaluated in a simulation study.     

Landscaping analyses of the ROC predictions of discrete-slots and signal-detection models of visual working memory

A fundamental issue concerning visual working memory is whether its capacity limits are better characterized in terms of a limited number of discrete slots (DSs) or a limited amount of a shared continuous resource. Rouder et al. (2008) found that a mixed-attention, fixed-capacity, DS model provided the best explanation of behavior in a change detection task, outperforming alternative continuous signal detection theory (SDT) models. Here, we extend their analysis in two ways: first, with experiments aimed at better distinguishing between the predictions of the DS and SDT models, and second, using a model-based analysis technique called landscaping, in which the functional-form complexity of the models is taken into account. We find that the balance of evidence supports a DS account of behavior in change detection tasks but that the SDT model is best when the visual displays always consist of the same number of items. In our General Discussion section, we outline, but ultimately reject, a number of potential explanations for the observed pattern of results. We finish by describing future research that is needed to pinpoint the basis for this observed pattern of results.     

Video Versus Reality: Infants' Attention and Affective Responses to Video and Live Presentations

Television viewing by infants has become very common, yet relatively little is known about infants' reactions to video compared to live events. We conducted two experiments in order to compare 9- or 10-month-old infants' behavioral reactions to live events and video presentations of those events. In the first experiment, infants' were shown video and live presentations shown consecutively; their affective reactions were consistent, indicating that video conveys meaning and emotional impact even in infancy. However, infants generally looked longer at, reached more to, showed more interest in, and exhibited more fear to the real stimuli. In a second experiment, infants were shown identical live and video events simultaneously; they looked much longer at the live presentations, further confirming that they distinguished live events from video and preferred to watch real events. Implications for infants' television viewing and the use of video images in infancy research are discussed.     

Cortisol,interleukins and S100B in delirium in the elderly

In independent studies delirium was associated with higher levels of cortisol, interleukin(IL)s, and S100B. The aim of this study was to simultaneously compare cortisol, IL-6, IL-8, and S100B levels in patients aged 65 years and older admitted for hip fracture surgery with and without delirium. Cortisol, IL-6, IL-8, and S100B were assayed in repeated blood samples. 120 patients (mean age 84 years, 62 patients with delirium) were included. Highest levels of IL-8 (27.1, 95% Confidence Interval (CI): 13.6–53.1 pg/ml) and cortisol (666, 95% CI: 475–859 nmol/L) were before delirium, but of IL-6 (84.3, 95% CI: 46.5–151.4 pg/mL) and S100B (0.18, 95% CI: 0.12–0.24 μg/L) during delirium. In multivariable analysis cortisol, LogIL-6, and LogS100B were significantly associated with delirium, but adjusted for pre-existing cognitive impairment, only LogS100B remained significantly associated. Cortisol, IL-6 and S100B may have a role in the pathogenesis of delirium, but S100B is the strongest independent marker.     

Decision criteria do not shift: Commentary on Mueller and Weidemann (2008)

The effects of base rates and payoffs on the shapes of rating receiver operating characteristic curves are inconsistent with the basic assumptions of signal detection theory (SDT), in particular the notion of a shifting decision criterion. Mueller and Weidemann (2008) propose that these unexpected phenomena are not due to problems with the decision- criterion construct but are instead due to two compounded effects: instability of the decision criterion across trials, and even greater instability in the flanking criteria that determine which confidence rating will be reported. There are several problems with the authors’ decision-noise hypothesis. First, even if their hypothesis about decision noise were taken for granted, the key feature of the ratings data that rejects the SDT model would remain a mystery. Second, the same violations of SDT that are exhibited in the ratings paradigm are also exhibited in the yes-no detection task when response time is substituted for confidence as a basis for analysis. Finally, the decision-noise hypothesis predicts that sensitivity will increase when one source of this variation-the response on a previous trial-is controlled for. This prediction was consistently violated in both the yes-no and ratings conditions of Mueller and Weidemann’s experiment. In an Addendum, we respond to Weidemann and Mueller’s (2008) reply to this Comment.     

Naughty or nice?: equity, gender and behavior

We review the debate over behavior, gender and classroom placement in ability groups for kindergartners. Using vignettes we vary children’s gender in three ways; male, female, or unspecified gender and also describe them as behaving well, average, or misbehaving. Our aim is to probe how much gender and behavior matter with respect to mock reading group placement. The design allows us to see if depicted behavior affects placement when every vignette child is described as possessing the same, average ability. After reading vignettes that vary gender and behavior, but not ability, respondents are directed to assign the portrayed child to a high, medium or low reading group. While depicted behavior strongly affects reading group placement, gender is not a significant predictor.

---

     

The origins of pictorial competence

Pictorial competence , which refers to the many factors involved in perceiving, interpreting, understanding, and using pictures, develops gradually over the first few years of life. Although experience is not required for accurate perception of pictures, it is necessary for understanding the nature of pictures. Infants initially respond to depicted objects as if they were real objects, and toddlers are remarkably insensitive to picture orientation. Only gradually do young children figure out the nature of pictures and how they are used.