Confirmation of Scientific Hypotheses as Relations

Summary  In spite of several attempts to explicate the relationship between a scientific hypothesis and evidence, the issue still cries for a satisfactory solution. Logical approaches to confirmation, such as the hypothetico-deductive method and the positive instance account of confirmation, are problematic because of their neglect of the semantic dimension of hypothesis confirmation. Probabilistic accounts of confirmation are no better than logical approaches in this regard. An outstanding probabilistic account of confirmation, the Bayesian approach, for instance, is found to be defective in that it treats evidence as a formal entity and this creates the problem of relevance of evidence to the hypothesis at issue, in addition to the difficulties arising from the subjective interpretation of probabilities. This essay purports to satisfy the need for a successful account of hypothesis confirmation by offering an original formulation based on the notion of instantiation of the relation urged by an hypothesis.     

An evidential support accumulation model of subjective probability

A model of cue-based probability judgment is developed within the framework of support theory. Cue diagnosticity is evaluated from experience as represented by error-free frequency counts. When presented with a pattern of cues, the diagnostic implications of each cue are assessed independently and then summed to arrive at an assessment of the support for a hypothesis, with greater weight placed on present than on absent cues. The model can also accommodate adjustment of support in light of the baserate or prior probability of a hypothesis. Support for alternatives packed together in a "residual" hypothesis is discounted; fewer cues are consulted in assessing support for alternatives as support for the focal hypothesis increases. Results of fitting this and several alternative models to data from four new multiple-cue probability learning experiments are reported.     

Bayesian estimation and testing of structural equation models

The Gibbs sampler can be used to obtain samples of arbitrary size from the posterior distribution over the parameters of a structural equation model (SEM) given covariance data and a prior distribution over the parameters. Point estimates, standard deviations and interval estimates for the parameters can be computed from these samples. If the prior distribution over the parameters is uninformative, the posterior is proportional to the likelihood, and asymptotically the inferences based on the Gibbs sample are the same as those based on the maximum likelihood solution, for example, output from LISREL or EQS. In small samples, however, the likelihood surface is not Gaussian and in some cases contains local maxima. Nevertheless, the Gibbs sample comes from the correct posterior distribution over the parameters regardless of the sample size and the shape of the likelihood surface. With an informative prior distribution over the parameters, the posterior can be used to make inferences about the parameters underidentified models, as we illustrate on a simple errors-in-variables model.We thank David Spiegelhalter for suggesting applying the Gibbs sampler to structural equation models to the first author at a 1994 workshop in Wiesbaden. We thank Ulf Böckenholt, Chris Meek, Marijtje van Duijn, Clark Glymour, Ivo Molenaar, Steve Klepper, Thomas Richardson, Teddy Seidenfeld, and Tom Snijders for helpful discussions, mathematical advice, and critiques of earlier drafts of this paper.     

样例类型与解释方式对初中生数学概率问题解决的效果

为考察样例类型与解释方式对初中生数学概率问题解决的促进作用,实验1随机选取初中生90名,比较正确样例组、正误样例组、对照组的学习效果,实验2随机选取另外90名初中生,比较有教学解释、有自我解释与无解释的正误样例组的即时与延时测试学习效果,研究发现:(1)正误样例学习效果显著好于正确样例;(2)有解释的正误样例学习效果显著好于无解释的正误样例;(3)与有教学解释的正误样例学习效果相比,有自我解释的正误样例学习效果显著且更持久。     

行为决策理论之父：纪念Edwards教授2周年忌辰

著名心理学家W ard Edwards对行为决策的学科形成与理论发展作出了巨大贡献。文章简要介绍了Edwards的生平,以主观期望效用模型和贝叶斯决策为核心回顾了Edwards的决策理论、主要成就及其对行为决策研究的学术贡献。文章并从行为决策领域的研究、科普、应用及人才培养等方面总结了中国行为决策研究的发展历程,以及Edwards的学术思想对中国行为决策研究的主要影响。文章展望了行为决策研究的前景,并呼吁更多的中国心理学工作者投身于行为决策研究。     

Likelihoodism,Bayesianism, and relational confirmation

Likelihoodists and Bayesians seem to have a fundamental disagreement about the proper probabilistic explication of relational (or contrastive) conceptions of evidential support (or confirmation). In this paper, I will survey some recent arguments and results in this area, with an eye toward pinpointing the nexus of the dispute. This will lead, first, to an important shift in the way the debate has been couched, and, second, to an alternative explication of relational support, which is in some sense a “middle way” between Likelihoodism and Bayesianism. In the process, I will propose some new work for an old probability puzzle: the “Monty Hall” problem. Thanks to the participants of the Philosophy, Probability, and Modeling (PPM) Seminar at the University of Konstanz (especially Stephan Hartmann, Franz Huber, Wolfgang Spohn, and Teddy Seidenfeld), for a very fruitful discussion of an early draft of this paper in July, 2004. Since then, discussions and correspondences with Prasanta Bandyopadhyay, Luc Bovens, Alan Hájek, Jim Hawthorne, Jim Joyce, Jon Kvanvig (and other participants of his “Certain Doubts” blog, which had a thread on a previous draft of this paper), Patrick Maher, Sherri Roush, Richard Royall, Elliott Sober, Dan Steel, and an anonymous referee of Synthese has been very valuable.     

CAT选题策略分类概述

选题是计算机化自适应测验（CAT）测试过程的关键环节,选题策略的目标是要达到较高的测量精度,同时也实现试题曝光率控制及其他测验目标的实现.本文根据选题策略的基本原理和衍生发展,将众多CAT选题策略分为五大选题策略系列：Fisher函数系列、K-LI函数系列、α分层系列、贝叶斯系列、b匹配系列;并根据测验目标（测验精度、试题曝光率控制、内容平衡、多条件约束）对这些选题策略进行了细分,并对CAT选题策略的选择思路进行归纳.     

De‐escalation after repeated negative feedback: emergent expectations of failure

Research on willingness to make marginal investments (e.g., the escalation and sunk cost literatures) has often focused on project completion decisions, such as the “radar‐blank plane.” This paper discusses a fundamentally different type of marginal investment decision, that of couples deciding whether to continue infertility treatment in the face of repeated failures. Two experiments based on this context show that when people face multiple independent chances to achieve a valued goal but are unsure about chances of success, premature quitting or “de‐escalation” is the norm. Repeated negative feedback appears to induce individuals to see each successive failure as more and more diagnostic. As a result, even a short series of failed attempts evokes beliefs that future attempts will also fail. These emergent expectations of failure, generated by causal attribution processes, associative learning, and/or discounting of ambiguous information, appear very compelling and induce people to forgo profitable marginal investments. Copyright © 2004 John Wiley & Sons, Ltd.     

Bayesian journey-to-crime modelling of single and multiple crime-type series in Baltimore County,MD

This study tested different Bayesian Journey-to-Crime (JTC) models on a sample group of 850 serial offenders apprehended in Baltimore County, MD from 1993 to 1997. In this research, Bayesian JTC models were being used to predict the home locations of the offenders. The sample group data included 133 assaults, 90 burglaries, 497 larcenies, 81 robberies, and 49 vehicle thefts. The main question this research aimed to answer was whether the addition of crimes of a different type to an existing crime series of a single type would result in more accurate and/or precise Bayesian JTC models. The standard practice by law enforcement has been to consider the same-type crime series only when modelling the anchor point of the offender. Similarly, in research, geographic profiles have been constructed exclusively with the same-type crime series. The results of this study clearly indicated that the inclusion of crimes of a different type into a single crime-type series will result in significantly more accurate and more precise Bayesian JTC models. In contrast, crime series with predominantly assault and burglary showed results that were inconclusive or indicated no significant differences. These results should encourage law enforcement agencies to re-evaluate their standard practice of constructing geographic profiles with only the same-type crime series. Copyright © 2009 John Wiley & Sons, Ltd.     

Global Model Analysis of Cognitive Variability

Residual fluctuations produced in typical experimental methodologies are examined as correlated noises. The effective range of the correlations was assessed by determining whether the decay over look-back time is better described as a power law or exponential. Both of these decay laws contain free parameters and it is argued that it is not possible to distinguish their models on the basis of simple measures of goodness-of-fit. Global analyses that evaluate models on the basis of how well they generalize are conducted. The models are examined in terms of three constructs that all bear on generalization: cross-validity, flexibility, and representativeness. Quantitative assessment of a large ensemble of data suggests that the correlations decay over time as a power law. The conclusion is that human residual fluctuation is a correlated fractal.