等级效标分数的概率神经网络预测方法研究

针对基于统计学的等级效标分数预测存在的问题,提出了运用概率神经网络进行等级效标分数预测的方法。在20种条件下进行了计算机模拟实验,包括测验分数为单变量和多变量,以及各种水平的白噪声干扰条件,结果表明在测验分数为多变量的情况下,或者在有白噪声干扰的条件下,运用概率神经网络方法可以比统计学方法更好地对等级效标分数进行预测。     

Dissimilarity cumulation theory and subjective metrics

We present a new mathematical notion, dissimilarity function, and based on it, a radical extension of Fechnerian Scaling, a theory dealing with the computation of subjective distances from pairwise discrimination probabilities. The new theory is applicable to all possible stimulus spaces subject to the following two assumptions: (A) that discrimination probabilities satisfy the Regular Minimality law and (B) that the canonical psychometric increments of the first and second kind are dissimilarity functions. A dissimilarity function Dab for pairs of stimuli in a canonical representation is defined by the following properties: (1) a≠b?Dab>0; (2) Daa=0; (3) If and , then ; and (4) for any sequence {a_nX_nb_n}_n∈N, where X_n is a chain of stimuli, Da_nX_nb_n→0?Da_nb_n→0. The expression DaXb refers to the dissimilarity value cumulated along successive links of the chain aXb. The subjective (Fechnerian) distance between a and b is defined as the infimum of DaXb+DbYa across all possible chains X and Y inserted between a and b.     

The origins of probabilistic inference in human infants

Reasoning under uncertainty is the bread and butter of everyday life. Many areas of psychology, from cognitive, developmental, social, to clinical, are interested in how individuals make inferences and decisions with incomplete information. The ability to reason under uncertainty necessarily involves probability computations, be they exact calculations or estimations. What are the developmental origins of probabilistic reasoning? Recent work has begun to examine whether infants and toddlers can compute probabilities; however, previous experiments have confounded quantity and probability—in most cases young human learners could have relied on simple comparisons of absolute quantities, as opposed to proportions, to succeed in these tasks. We present four experiments providing evidence that infants younger than 12 months show sensitivity to probabilities based on proportions. Furthermore, infants use this sensitivity to make predictions and fulfill their own desires, providing the first demonstration that even preverbal learners use probabilistic information to navigate the world. These results provide strong evidence for a rich quantitative and statistical reasoning system in infants.     

Ownership Status and the Representation of Assets of Uncertain Value: The Balloon Endowment Risk Task (BERT)

Owners tend to overvalue possessions relative to non‐owners: a phenomenon known as the endowment effect. In three experiments, using markets for goods of uncertain value, we investigated whether this can be partly attributed to misperceiving an asset's profitability or to uncertainty about a good's utility. To test our hypotheses, we devised the Balloon Endowment Risk Task, in which participants can sell or buy their right to participate in the Balloon Analogue Risk Task. Once purchased/retained, a virtual balloon is pumped to accrue money, which is lost if the balloon bursts. Participants first learn about the risky asset (balloon) by observing others playing the Balloon Analogue Risk Task before they enter the market. In Experiment 1, we replicated the endowment effect; yet, owners and non‐owners predicted pumping the same number of times and subsequently did so when given that opportunity. In Experiments 2 and 3, the level of uncertainty about the balloon's profitability was manipulated by modifying the number of bursts that participants viewed in the initial learning stage. When more diagnostic information was provided, making the average burst point easier to estimate and reducing value uncertainty and increasing confidence in valuation, the endowment effect diminished although mean estimates of the average burst point did not differ between owners and non‐owners. Thus, endowment effects were partly attributable to value uncertainty but could not be explained by owners and non‐owners having divergent perceptions of the asset's payoff distribution. Copyright © 2014 John Wiley & Sons, Ltd.     

The Equiprobability Bias from a Mathematical and Psychological Perspective

The equiprobability bias (EB) is a tendency to believe that every process in which randomness is involved corresponds to a fair distribution, with equal probabilities for any possible outcome. The EB is known to affect both children and adults, and to increase with probability education. Because it results in probability errors resistant to pedagogical interventions, it has been described as a deep misconception about randomness: the erroneous belief that randomness implies uniformity. In the present paper, we show that the EB is actually not the result of a conceptual error about the definition of randomness. On the contrary, the mathematical theory of randomness does imply uniformity. However, the EB is still a bias, because people tend to assume uniformity even in the case of events that are not random. The pervasiveness of the EB reveals a paradox: The combination of random processes is not necessarily random. The link between the EB and this paradox is discussed, and suggestions are made regarding educational design to overcome difficulties encountered by students as a consequence of the EB.     

Waiting When Both Certainty and Magnitude Are Increasing: Certainty Overshadows Magnitude

In everyday decision making, people often face decisions with outcomes that differ on multiple dimensions. The trade‐off in preferences between magnitude, temporal proximity, and probability of an outcome is a fundamental concern in the decision‐making literature. Yet, their joint effects on behavior in an experience‐based decision‐making task are understudied. Two experiments examined the relative influences of the magnitude and probability of an outcome when both were increasing over a 10‐second delay. A first‐person shooter video game was adapted for this purpose. Experiment 1 showed that participants waited longer to ensure a higher probability of the outcome than to ensure a greater magnitude when experienced separately and together. Experiment 2 provided a precise method of comparing their relative control on waiting by having each increase at different rates. Both experiments revealed a stronger influence of increasing probability than increasing magnitude. The results were more consistent with hyperbolic discounting of probability than with cumulative prospect theory's decision weight function. Copyright © 2014 John Wiley & Sons, Ltd.     

四参数Logistic模型潜在特质参数的Warm加权极大似然估计

本文以四参数Logistic(4-parameter Logistic,4PL)模型为研究对象,根据Warm的加权极大似然估计技巧,提出了4PL模型潜在特质参数的加权极大似然估计方法,并借助模拟研究对加权极大似然估计的性质进行验证。研究结果表明,与通常的极大似然估计和后验期望估计相比,加权极大似然估计的偏差(bias)明显减小,并且具有良好的返真性能。此外,在测试的长度较短和项目的区分度较小的情况下,加权极大似然估计依然保持了良好的统计性质,表现出更加显著的优势。     

The Appeal to Expert Opinion: Quantitative Support for a Bayesian Network Approach

The appeal to expert opinion is an argument form that uses the verdict of an expert to support a position or hypothesis. A previous scheme‐based treatment of the argument form is formalized within a Bayesian network that is able to capture the critical aspects of the argument form, including the central considerations of the expert's expertise and trustworthiness. We propose this as an appropriate normative framework for the argument form, enabling the development and testing of quantitative predictions as to how people evaluate this argument, suggesting that such an approach might be beneficial to argumentation research generally. We subsequently present two experiments as an example of the potential for future research in this vein, demonstrating that participants' quantitative ratings of the convincingness of a proposition that has been supported with an appeal to expert opinion were broadly consistent with the predictions of the Bayesian model.     

非参数认知诊断方法下诊断结果的概率化表征

非参数认知诊断分类方法非常适合课堂评估,其诊断结果采用0-1形式而缺乏概率化表征,不能精细地区分被试属性掌握程度的差异或变化,还缺乏可用于评价真实测验分类结果的信度和效度指标。要刻画被试属性掌握程度的差异,首要的问题是要为非参数认知诊断方法提供一种可以量化属性掌握概率的方法。针对此问题,基于二项分布和玻尔兹曼分布提出非参数认知诊断方法下诊断结果的概率化表征方法,并用于构建分类准确性和分类一致性指标。模拟研究与实测数据分析结果显示：概率化表征方法与非参数认知诊断方法的分类结果高度一致;概率化表征方法与认知诊断模型所得的属性掌握概率十分接近;概率化表征方法所得的属性（模式）掌握概率可用于计算属性（模式）分类准确性和分类一致性指标,在实际测验情景下可作为信度和效度指标,评价诊断结果的重测一致率和判准率。