Generalization gradients in a navigation task with rats

In two experiments in a Morris pool, rats were trained to find a hidden platform which was located in a specific position in relation to two objects, B and F, which were presented together, one in front of the other. One object, B, was just above the platform (a beacon for the platform, the critical object) while the second object, F, was above the edge of the pool (the frame of reference). Then the rats received test trials, without the platform, in which B was presented in different positions in relation to F (i.e., in relation to its original position). In the two experiments the test results showed a generalization gradient as a function of the relative distance of the two objects: more time searching in the B segment, where the platform should have been, when B was in the original position (i.e., in front of F), which decreased symmetrically with distance of B from F. The present experiments show for the first time generalization gradients with rats across spatial locations when working with a navigation task.     

Spontaneous Response Generalization During Whole Word Instruction: Reading to Spell and Spelling to Read

The bulk of the existing literature emphasizes the use of phonemic/phonetic based instruction to enhance generalization between reading and written language. However, phonetically irregular words are common in English and may require the use of whole word approaches. This study examined generalization between from reading to spelling and from spelling to reading following whole word based instruction using a delayed prompt procedure. The total instructional time was equated across conditions. Students exhibited substantial but incomplete generalization with the degree of generalization varying somewhat across responses and students. The most striking finding was that oral spelling instruction in which the printed word was neither presented nor produced resulted in substantial generalization to reading. These data illustrate a model for examining response generalization in instructional contexts.     

Teaching Phoneme Segmenting and Blending as Critical Early Literacy Skills: An Experimental Analysis of Minimal Textual Repertoires

The purpose of the current study was to examine blending and segmenting of phonemes as an instance of small, textual response classes that students learn to combine to produce whole word reading. Using an A/B/A/B design, a phoneme segmenting and blending condition that included differential reinforcement for response classes at the level of phonemes was compared to a control condition which was equated for differential reinforcement of reading words and opportunities to respond. The critical difference between conditions was the size of the responses that were brought under stimulus control (phonemes versus whole words). Findings clearly supported the superiority of the phoneme blending treatment condition over the control condition in producing generalized increases in word reading. The results are discussed in terms of the behavioral mechanisms that govern early literacy behaviors and the essential role that targeting measured increases in academic responses plays in furthering our understanding of how to improve the analysis and instruction of students who need to learn these important skills.     

Sitting on the word “chair”: Behavioral support, contextual cues, and the literal use of symbols

Recent work from an embodied-cognition perspective suggests that symbolic understanding involves bodily actions. Indeed, laboratory evidence and cultural phenomena such as magic rituals and symbolic aggression show that the behaviors evoked by a word and its referent can be quite similar to each other. In other circumstances, however, words and objects fail to display the expected degree of functional equivalence: Although we regularly sit on chairs, for example, we normally do not sit on the word “chair.” In two experiments we evaluated whether functional equivalence between word and object depended on behavioral support and contextual stimuli, as suggested by some Pavlovian views of symbolic performance. Children spent more or less time seated on a token of the word “chair,” depending on supports for sitting and background stimuli. This study may help to understand the determinants of response frequency in symbolic understanding.     

Measuring self-serving cognitive distortions: A meta-analysis of the psychometric properties of the How I Think Questionnaire (HIT)

The How I Think Questionnaire (HIT; Barriga et al., 2001) is a self-report measure of self-serving cognitive distortions. This study used meta-analytic methods to analyse: (i) the reliability of the HIT scores across samples; (ii) the variability in reliability estimates; and (iii) the convergent and criterion-group validity of the HIT scores. Results showed an excellent reliability of the total HIT scores: mean alpha = .93, 95% CI [.92, .94]; k = 29, N = 8,186. Heterogeneity of total HIT was mainly explained by type of population. The reliabilities of the four subscale scores were also adequate. Moreover, the HIT has strong convergent validity and is able to discriminate between offenders and control participants. Limitations of the current literature are discussed.     

Feature selection for inductive generalization

Judging similarities among objects, events, and experiences is one of the most basic cognitive abilities, allowing us to make predictions and generalizations. The main assumption in similarity judgment is that people selectively attend to salient features of stimuli and judge their similarities on the basis of the common and distinct features of the stimuli. However, it is unclear how people select features from stimuli and how they weigh features. Here, we present a computational method that helps address these questions. Our procedure combines image-processing techniques with a machine-learning algorithm and assesses feature weights that can account for both similarity and categorization judgment data. Our analysis suggests that a small number of local features are particularly important to explain our behavioral data.     

成年人分类概括力水平的年龄差异及其与记忆的关系

采用威斯康星卡片分类测验（Ｗｉｓｃｏｎｓｉｎｃａｒｄｓｏｒｔｉｎｇｔｅｓｔ,ＷＣＳＴ）对青年组（２０—２５岁）、老年组（６０—６５岁）和老老年组（７０—７５岁）共９０例被试进行分类概括力水平测试,并在结束测试时询问被试与测试有关的几个问题。结果显示：青年组分类概括力水平明显高于老年组;坚持性错误数、非坚持性错误数和总错误数以及坚持性反应数,青年组明显少于老年组;完成作业平均用时间,青年组明显比老年组短;老老年组ＷＣＳＴ多数项目的成绩明显高于老年组;分类概括力水平与记忆密切相关;某些元认知项目与分类概括力水平表现出一致的趋势。     

TRAINING AND GENERALIZATION OF AFFECTIVE BEHAVIOR DISPLAYED BY YOUTH WITH AUTISM

The purpose of this study was to teach contextually appropriate affective behavior to 4 youths with autism. Treatment consisted of modeling, prompting, and reinforcement introduced in a multiple baseline design across response categories of affective behavior. During treatment, verbal praise and tokens were delivered contingent on appropriate affective responding during training trials. Modeling and verbal prompting were used as correction procedures. Each youth received treatment in either three or four response categories. Treatment systematically increased responding within the response categories for all 4 participants, with effects being specific to the affective response categories under treatment. Treatment effects occurred across untrained scenarios, therapists, time, and settings, suggesting that generalization had occurred.     

EPQ信度概化的跨文化比较及其启示

对中国7种主要的心理学杂志,于1998-2008年间发表的与EPQ有关的研究报告或论文做信度概化分析,并与Caruso等人对其他国家的EPQ信度概化分析结果作比较。分析结果表明：中外心理量表使用者都存在严重的“信度引入”的状况;分量表的分数的标准差是信度系数最重要的预测变量;但其他预测变量有所不同。提供的启示是：使用心理量表时一定要报告当前样本的信度系数;不符合心理量表特性要求而增加项目,不一定能提高测验结果的信度。     

资格认证测验的信度估计及其特征分析

资格认证测验属于典型的标准参照测验,在国内已得到普遍应用,但在报告心理测量学指标时很少提到标准参照模式的信度估计指标。该文归纳了标准参照测验信度估计的指标体系,分析讨论了适用于资格认证测验的信度估计及其与测验长度、分界标准分布、样本同质性的关系与特征。