Likelihood-Based Clustering of Meta-Analytic SROC Curves

Meta-analysis of diagnostic studies experience the common problem that different studies might not be comparable since they have been using a different cut-off value for the continuous or ordered categorical diagnostic test value defining different regions for which the diagnostic test is defined to be positive. Hence specificities and sensitivities arising from different studies might vary just because the underlying cut-off value had been different. To cope with the cut-off value problem interest is usually directed towards the receiver operating characteristic (ROC) curve which consists of pairs of sensitivities and false-positive rates (1-specificity). In the context of meta-analysis one pair represents one study and the associated diagram is called an SROC curve where the S stands for “summary”. In meta-analysis of diagnostic studies emphasis has traditionally been placed on modelling this SROC curve with the intention of providing a summary measure of the diagnostic accuracy by means of an estimate of the summary ROC curve. Here, we focus instead on finding sub-groups or components in the data representing different diagnostic accuracies. The paper will consider modelling SROC curves with the Lehmann family which is characterised by one parameter only. Each single study can be represented by a specific value of that parameter. Hence we focus on the distribution of these parameter estimates and suggest modelling a potential heterogeneous or cluster structure by a mixture of specifically parameterised normal densities. We point out that this mixture is completely nonparametric and the associated mixture likelihood is well-defined and globally bounded. We use the theory and algorithms of nonparametric mixture likelihood estimation to identify a potential cluster structure in the diagnostic accuracies of the collection of studies to be analysed. Several meta-analytic applications on diagnostic studies, including AUDIT and AUDIT-C for detection of unhealthy alcohol use, the mini-mental state examination for cognitive disorders, as well as diagnostic accuracy inspection data on metal fatigue of aircraft spare parts, are discussed to illustrate the methodology.     

The Informative Analysis of Individual Trend Curves

Additive and non-additive models for an individual trend curve are examined, and five methods for fitting these to a set of individuals are described. It is suggested that classical fitting methods are more informative than latent curve methods, and commonly preferable. A limited study of the effect of time-structure is reported, and results on the relationship between a non-additive model and the approximating additive model are given.     

关于强迫折返的可计算曲线

平面曲线是指单位闭区间在连续函数映射下的影像。这样的函数叫做曲线的参数化。如果一条曲线有一个单调的参数化函数,该曲线是简单曲线。如果一条曲线有可计算的参数化函数,则相应的曲线是可计算的。Gu、LutzHMayordomo最近证明了,有些可计算的简单曲线没有可计算的参数化函数。这样的曲线叫做强迫折返的。本文探讨的是强迫折返的次数对曲线复杂性的影响。     

Schedule thinning following communication training: using competing stimuli to enhance tolerance to decrements in reinforcer density

Functional communication training (FCT) is a widely used treatment for individuals with developmental disabilities who exhibit severe behavior problems. One inherent challenge of employing FCT as a treatment in the community is that reinforcement for appropriate communication cannot always be immediate or even possible in some circumstances. Of the few studies that have incorporated some form of schedule thinning for communication, most have reported disruption of communication or increases in problem behavior when the schedule of reinforcement is thinned. In the current study, we compared FCT with extinction to FCT with extinction and access to competing stimuli. After conducting a functional analysis, a competing stimulus assessment was performed to identify stimuli that produce reinforcement that ostensibly competes with reinforcement that maintains problem behavior. It was hypothesized that FCT with competing stimuli would result in more stable reductions in problem behavior during schedule thinning, which would ultimately result in quicker achievement of the treatment goal (low levels of problem behavior under the terminal reinforcement schedule for communication) than FCT without competing stimuli. Results confirmed this hypothesis.     

Measurement of ambiguity tolerance

Summary: Presented definitions for the construct of ambiguity tolerance. The measun: of ambiguity tolerance (MAT-50) had high internal reliability (r = .88) and high test-retest reliability (r = .86) over a 10-to-12 week period. A content analysis of the measure and a subjective imalysis by 20 graduate students indicated adequate content validity. A multivariate comparison with two other ambiguity measures, two rigidity measures, and a short dogmatism measure provided strong evidence for criteria-related validity. Finally, four independent empirical studies showed good construct validity.     

Predicting Growth Curves of Externalizing Behavior Across the Preschool Years

Hierarchical linear modeling was used to examine 2 facets of externalizing behavior: its level at age 6 and its rate of change from age 2 to 6 among 299 boys from low-income families. As for age-6 level of externalizing behavior, maternal depressive symptoms, maternal acceptance of child behavior, parental conflict, and the interaction between maternal acceptance and maternal depressive symptoms were each uniquely associated with this outcome. Regarding the interaction, maternal acceptance was associated with externalizing behavior primarily when maternal depressive symptoms were low, and maternal depressive symptoms were related to externalizing behavior only when maternal acceptance was high. Externalizing behavior gradually decreased between ages 2 and 6; interactions between infant negative emotionality and maternal depressive symptoms and between parental conflict and maternal acceptance were associated with rate of change. Plots suggested a protective effect of low infant negative emotionality when maternal depressive symptoms were high and a larger effect of maternal acceptance when parental conflict was low compared to when it was high. Results add to a small but growing literature regarding predictors of continuity and change in early child externalizing behavior.     

Political tolerance and coming to psychological closure following the September 11, 2001, terrorist attacks: an integrative approach

This study tested hypotheses generated from an integrative model of political tolerance that derived hypotheses from a number of different social psychological theories (e.g., appraisal tendency theory, intergroup emotion theory, and value protection models) to explain political tolerance following the September 11, 2001, terrorist attacks. A national field study (N = 550) found that immediate post attack anger and fear had different implications for political tolerance 4 months later. The effects of anger on political tolerance were mediated through moral outrage and outgroup derogation, whereas the effects of fear on political tolerance were mediated through personal threat, ingroup enhancement, and value affirmation. Value affirmation led to increased political tolerance, whereas moral outrage, outgroup derogation, ingroup enhancement, and personal threat led to decreased political tolerance. Value affirmation, moral outrage, and outgroup derogation also facilitated post-9/11 psychological closure and increased psychological closure led to greater political tolerance.     

Sensitivity to low doses of ethanol and pentobarbital in mice selected for sensitivity to hypnotic doses of ethanol.

These experiments assessed sensitivity to low doses of ethanol and pentobarbital in mice that had been selectively bred with respect to ethanol sleep time (the length of time an animal remains on its back following a hypnotic dose of ethanol). The hypothesis under investigation was that short-sleep (SS) mice might be more sensitive than long-sleep (LS) mice to excitatory effects produced by low doses of depressants. In support of this hypothesis SS mice were more active in an open-field test after ethanol than were LS mice. The lines did not differ in performance on a rotating-rod apparatus after these same doses of ethanol, suggesting that the difference in open-field activity was not attributable to a greater impairment of locomotor activity in LS mice. A similar difference in the open-field activity of the selected lines was observed with pentobarbital.     

The logic of linear tolerance

A nonempty sequence T₁,...,T_n of theories is tolerant, if there are consistent theories T ₁ ⁺ ,..., T _n ⁺ such that for each 1 i n, T _i ⁺ is an extension of T_i in the same language and, if i n, T _i ⁺ interprets T _i+1 ⁺ . We consider a propositional language with the modality , the arity of which is not fixed, and axiomatically define in this language the decidable logics TOL and TOL. It is shown that TOL (resp. TOL) yields exactly the schemata of PA-provable (resp. true) arithmetical sentences, if (A₁,..., A_n) is understood as (a formalization of) PA+A₁, ..., PA+A_n is tolerant.     

Extinction of morphine analgesic tolerance

It has been suggested that the analgesic effect of morphine becomes attenuated over the course of successive administrations by a conditional, compensatory, hyperalgesic response elicited by the administration procedure, thus accounting (in part) for analgesic tolerance. On the basis of this associative model of tolerance, it would be predicted that established tolerance would be extinguished by placebo sessions. In experiments of apparently similar design (but conducted by different experimenters in different laboratories), data both confirming and refuting this prediction have previously been published. The present experiment, conducted jointly by the experimenters who reported the divergent findings, was designed to determine the reasons for the different results. It was found that placebo sessions do consistently attenuate morphine analgesic tolerance. Such extinction is not limited to the experimenter, drug preparation, rat strain, or apparatus used in the original, successful demonstration of the phenomenon, but rather is also demonstrable under conditions similar to those used in subsequent experiments which failed to demonstrate extinction of tolerance. Results of the present experiment suggest that the failures to demonstrate extinction of tolerance were attributable to insufficient extinction training.     

State-dependent recall decrements with moderate doses of alcohol

Thirty-two adult volunteers were asked to memorize a simplified geographical map, visually displayed, together with an auditorially presented 19-item set of instructions regarding a particular route. During this session, half the subjects were sober and half were under the effects of a moderate dose of alcohol (mean blood alcohol concentration = 81mg/100ml). Twenty-four hours later they were tested under the same or different conditions. Learning performance on Day 1 was unaffected by alcohol. Learning transfer (tested on Day 2) was equally good when subjects were intoxicated during both sessions or were sober during both sessions. There were significant decrements in recall on Day 2, when subjects were in a different drug state. This ‘dissociation of learning’ was found to be symmetrical in that the recall decrement was the same whether the initial learning was acquired in a sober or a drugged state.