认知诊断测验蓝图的设计

通常认为由属性和项目关联阵（即Q矩阵）的列对应的项目充任认知诊断测验中行为样本,其实这种做法不能有效防止理想反应模式的误判。如在测验之前便可确定欲测之属性及层级关系,找到可达阵,可证明可达阵的各个列对应的项目类在认知诊断测验中必不可少,否则在理想反应模式下就一定有一些被试会被误判。本文给出充分必要Q矩阵的概念,以区别Tatsuoka(1995,2009) 讨论过的充分Q矩阵概念。充分必要Q矩阵才能有效指导测验的编制。     

Searching for Evidence of Transfer between Drug Facilitators

Four autoshaping experiments with pigeons examined the transfer of facilitation between drug stimuli. In Experiment 1 the presence and absence of methadone or phencyclidine signaled which of two keylight CSs were followed by grain (US). A different keylight CS was reinforced in the absence of the drugs. The drugs facilitated responding to the CS with which they were trained but did not facilitate responding to the CS reinforced in the presence of the other drug. Thus, transfer of facilitation across drug stimuli was not obtained. In Experiment 2 one of the drug facilitators was replaced by a tone. No evidence of transfer of facilitative function was obtained between the tone and the remaining drug facilitator. In Experiment 3 transfer was again not found using a design similar to that used in Experiment 1 except that the drugs did not simultaneously signal nonreinforcement of a keylight CS. In Experiment 4 the procedures of Experiment 1 were employed except that drug facilitators were replaced by distinct visual stimuli surrounding the response key. Clear evidence of transfer between the visual facilitators was obtained. The lack of transfer of modulatory control by drug stimuli in Experiments 1-3, coupled with its occurrence in Experiment 4 and previous studies, suggests that stimulus control by drugs may differ in important ways from stimulus control by conventional stimuli.     

从《太玄》看扬雄的人性论思想

一般讲扬雄的人性论,都以《法言》为依据,认为他是“善恶混”派的代表.并认为他的人性论缺乏内在超越的根据,比较接近于荀子。本文认为《太玄》中蕴涵丰富的人性论思想,研究扬雄的人性论,应以《太玄》为基础。扬雄的人性论本于其天道观,有很强的超越性,扬氏以五行为架构,提出了非常具体的超越途径,其思想接近于性善说.     

Repeated,within‐session resurgence

Resurgence is a reliable, transient effect that only occasionally is replicated more than once within a single experiment or subject. In the present experiments, within‐session resurgence was generated repeatedly by dividing individual sessions into three phases (Training, Alternative‐Reinforcement, and Resurgence‐Test). In Experiments 1 and 2, resurgence reliably occurred in most of the 22‐30 daily sessions when responding was reinforced on, respectively, fixed‐ and variable‐interval schedules. Resurgence magnitude and duration did decrease across replications for some subjects, but not for others. To examine the utility of the procedure in studying the effects of an independent variable on resurgence, in Experiment 3 the effects of rich and lean baseline and alternative reinforcement rates on resurgence were compared. The target response was eliminated more rapidly, resurgence occurred more often, and usually was greater following rich alternative reinforcement rates. Resurgence was of greater magnitude when the baseline reinforcement rate was relatively lean compared to the alternative reinforcement rate. These experiments provide a reliable method for generating resurgence within individual sessions, instead of across multiple‐session conditions, that can be repeated over many successive sessions.     

Reinforcer quality matters: A test of the Mathematical Principles of Reinforcement with domestic hens (Gallus gallus domesticus)

This study evaluated the ability of Killeen's (1994) Mathematical Principles of Reinforcement to account for the effects of changes in reinforcer quality on hens' rates of responding on fixed‐ratio schedules. Hens were trained to peck a key on a fixed‐ratio schedule of reinforcement and then experienced an ascending series of ratio values in two separate conditions. In different conditions, the food reinforcer was either wheat or puffed wheat. Response rates initially increased with increases in ratio requirement before eventually decreasing at larger ratios. Quantitative fits of the model accounted for the data well. The fits revealed that different foods were systematically associated with changes in the specific activation parameter, a, and these were consistent with previous reports of preference for those food items.     

Hierarchical resurgence

In resurgence, conventionally a target response is trained and then extinguished while some alternative response is reinforced. In the most common procedure, when the latter is extinguished, the former resurges. The present experiments examined resurgence after two responses were trained sequentially and subsequently extinguished. In Experiments 1 and 2, keypecking to one key was trained and then extinguished as keypecking to a different key was trained then later extinguished. In both experiments, regardless of the spatial location of the different keys, the last‐trained response resurged before the first‐trained one. The results were replicated in Experiment 3 where reinforcement rate of the first‐trained response was four times that of the second‐trained response. The results in conjunction with earlier experiments suggest that resurgence occurs hierarchically, although whether more or less recently trained target responses resurge first or later may depend on both current and historical variables. The results also raise questions about the interpretation of responding on a control key that sometimes is included in resurgence experiments to isolate resurgence from extinction‐induced responding.     

Social influence in investigative interviews: The effects of authority on informational elements produced in interviews and written statements

This study examined the effect of a principle of social influence—authority—on the informational elements in an investigative interview. Community participants told the truth or lied about a mock crime in a high or low authority context. Informational elements were coded from their oral responses in the interviews and written responses in a statement prior to the interviews. Rapport was rated by the interviewers and participants. The authority condition produced effects on the informational elements in both the interviews and written statements but mainly for truth tellers. Rapport also had direct effects on the informational elements but authority did not affect rapport and rapport did not mediate the effect of authority on the informational elements. These findings had theoretical, empirical, and practical implications.     

基于属性掌握概率的认知诊断计算机化自适应测验选题策略

在认知诊断计算机化自适应测验(CD-CAT)中, 被试对每个属性的掌握概率更直接地反映了被试能力的当前估计值。因此, 基于被试的属性掌握概率来构建选题策略, 选择最能改变被试属性掌握概率的题目作为下一个测验项目, 这应该是一个值得尝试的方案。本文借鉴已有相关研究的数据生成模式进行探索, 模拟实验结果表明：假设属性间相互独立,在定长(长度为16)、变长(长度为16或后验属性掌握模式概率达到0.8)以及短测验(长度分别为4、6、8、10)的情况下, 基于属性掌握概率的选题策略PPWKL和PHKL有较好的分类准确率, 在题目曝光率, 题库使用均匀性等方面也有较好的表现; 与研究较多的PWKL、HKL等策略相比, 也略有优势; 当属性间存在不同程度的相关时, 在定长、变长以及较短的测验条件下, 基于PHKL和MI的测验对知识状态估计精度较好, 基于PPWKL和PHKL的测验综合表现占优。     

The effects of reinforcer magnitude in the preceding and upcoming ratios on between‐ratio pausing in multiple,mixed, and single fixed‐ratio schedules

Hens responded under multiple fixed‐ratio schedules with equal response requirements and either a 1‐s or a 6‐s reinforcer. Upcoming reinforcer size was indicated by key color. Components were presented in a quasirandom series so that all four component transitions occurred. Postreinforcement pauses were affected by the upcoming and preceding reinforcer size, with longer pauses after large reinforcers followed by small reinforcers than when followed by large ones, and longer pauses after small reinforcers that were followed by small reinforcers rather than large ones. Pauses increased with fixed‐ratio size and the effects of reinforcer size were larger the larger the ratio. When reinforcer size was not signaled—mixed fixed‐ratio schedules—pauses were shorter after small than after large reinforcers. Signalling the upcoming reinforcer attenuated the effect of the previous reinforcer size on pause duration when small was followed by small and when either small or large by large, but enhanced the effect when large was followed by small. There was no effect of reinforcer size on pause duration when single fixed‐ratio schedules were arranged. The effects of reinforcer size on pauses depends on the size and range of the fixed ratios as well as the exact procedures used in the study.