A Bayesian Semiparametric Latent Variable Model for Mixed Responses

In this paper we introduce a latent variable model (LVM) for mixed ordinal and continuous responses, where covariate effects on the continuous latent variables are modelled through a flexible semiparametric Gaussian regression model. We extend existing LVMs with the usual linear covariate effects by including nonparametric components for nonlinear effects of continuous covariates and interactions with other covariates as well as spatial effects. Full Bayesian modelling is based on penalized spline and Markov random field priors and is performed by computationally efficient Markov chain Monte Carlo (MCMC) methods. We apply our approach to a German social science survey which motivated our methodological development. We thank the editor and the referees for their constructive and helpful comments, leading to substantial improvements of a first version, and Sven Steinert for computational assistance. Partial financial support from the SFB 386 “Statistical Analysis of Discrete Structures” is also acknowledged.     

Evidence of the origin of specific spontaneous head turns during intertrial intervals

Direction and the frequency of spontaneous head movements during the ITIs following forward and backward paired trials were compared to an acquisition of a conditioned orienting (alpha) response directed to the side of the tone source. The head movements were analyzed from video recordings using classification of head turns to preferred and to nonpreferred directions. The results showed a significant increase in the alpha responses during the forward paired conditioning to the preferred direction and rapid extinction during the subsequent backward conditioning sessions. Spontaneous head movements during the ITIs increased to the same preferred direction as the conditioned alpha responses. The results of this experiment suggest that the response initially elicited by the CS can later appear as “spontaneous,” instrumental behavior, the form and the nature of which is determined by the characteristics of the conditioned alpha response developing as a result of classical conditioning.     

Models for measurement,precision, and the nondichotomization of graded responses

It is common in educational, psychological, and social measurement in general, to collect data in the form of graded responses and then to combine adjacent categories. It has been argued that because the division of the continuum into categories is arbitrary, any model used for analyzing graded responses should accommodate such action. Specifically, Jansen and Roskam (1986) enunciate ajoining assumption which specifies that if two categoriesj andk are combined to form categoryh, then the probability of a response inh should equal the sum of the probabilities of responses inj andk. As a result, they question the use of the Rasch model for graded responses which explicitly prohibits the combining of categories after the data are collected except in more or less degenerate cases. However, the Rasch model is derived from requirements of invariance of comparisons of entities with respect to different instruments, which might include different partitions of the continuum, and is consistent with fundamental measurement. Therefore, there is a strong case that the mathematical implication of the Rasch model should be studied further in order to understand how and why it conflicts with the joining assumption. This paper pursues the mathematics of the Rasch model and establishes, through a special case when the sizes of the categories are equal and when the model is expressed in the multiplicative metric, that its probability distribution reflects the precision with which the data are collected, and that if a pair of categories is collapsed after the data are collected, it no longer reflects the original precision. As a consequence, and not because of a qualitative change in the variable, the joining assumption is destroyed when categories are combined. Implications of the choice between a model which satisfies the joining assumption or one which reflects on the precision of the data collection considered are discussed.     

Departure from normal assumptions: A promise for future psychometrics with substantive mathematical modeling

Normal assumptions have been used in many psychometric methods, to the extent that most researchers do not even question their adequacy. With the rapid advancement of computer technologies in recent years, psychometrics has extended its territory to include intensive cognitive diagnosis, etcetera, and substantive mathematical modeling ha become essential. As a natural consequence, it is time to consider departure from normal assumptions seriously. As examples of models which are not based on normality or its approximation, the logistic positive exponent family of models is discussed. These models include the item task complexity as the third parameter, which determines the single principle of ordering individuals on the ability scale.     

Conditional discrimination in mentally retarded adults: the development of generalized skills.

The development of generalized conditional discrimination skills was examined in adults with retardation. Two subjects with histories of failure to acquire arbitrary matching under trial-and-error procedures were successful under procedures that trained one or more prerequisite skills. The successive discrimination between the sample stimuli was established by training the subjects to name the stimuli. The simultaneous discrimination between the comparison stimuli was established using either (a) standard simple discrimination training with reversals or (b) a procedure in which each of the two sample-comparison relations in the conditional discrimination was presented in blocks of trials, with the size of the blocks decreasing gradually until sample presentation was randomized. The amount of prerequisite training required varied across subjects and across successive conditional discriminations. After acquiring either two or three conditional discriminations with component training, both subjects learned new conditional discriminations under trial-and-error procedures. In general, each successive conditional discrimination was acquired more rapidly. Tests showed that conditional responding had become a generalized skill. Symmetry was shown for almost all trained relations. Symmetry trial samples were ultimately named the same as the stimuli to which they were related in training.     

The effect of question type on resistance to misinformation about present and absent details

The typical misinformation effect shows that accuracy is lower for details about which people received misleading compared to non-misleading (control) information. In two experiments, we examined the misinformation effect for non-witnessed details (i.e., absent). Three question types introduced control, misleading, and absent details (closed, closed-detailed, and open questions) about a mock burglary video. On this misinformation test, participants' reports of absent details were less accurate than control details only when they were introduced using open questions. Misinformation effects in a subsequent recognition test were present for misleading details in both experiments, but for absent details only in Experiment 2. Experiment 2 also revealed that participants who avoided answering open questions containing misleading and absent details had more accurate memories for these details on the subsequent recognition test than participants who answered these questions. In both experiments, confidence was lowest for absent details. Implications for theory and practice are discussed.     

高血压患者应激反应特点及行为人格的关系研究

为了解高血压患者心血管反应的特征及与健康人群的差异,使用实验和问卷调查的方法对高血压患者和健康居民进行研究。结果显示,两组在任务期和恢复期的心血管反应差异显著（P〈0．05）;病例组的健康实践行为和消极应对方式水平较高,积极应对方式有待提升;健康实践行为和恢复期舒张压呈负相关,积极应对方式和任务期收缩压呈正相关（P〈0...     

Phonological,visual, and semantic coding strategies and children's short-term picture memory span

Three experiments addressed controversies in the previous literature on the development of phonological and other forms of short-term memory coding in children, using assessments of picture memory span that ruled out potentially confounding effects of verbal input and output. Picture materials were varied in terms of phonological similarity, visual similarity, semantic similarity, and word length. Older children (6/8-year-olds), but not younger children (4/5-year-olds), demonstrated robust and consistent phonological similarity and word length effects, indicating that they were using phonological coding strategies. This confirmed findings initially reported by Conrad (1971), but subsequently questioned by other authors. However, in contrast to some previous research, little evidence was found for a distinct visual coding stage at 4 years, casting doubt on assumptions that this is a developmental stage that consistently precedes phonological coding. There was some evidence for a dual visual and phonological coding stage prior to exclusive use of phonological coding at around 5–6 years. Evidence for semantic similarity effects was limited, suggesting that semantic coding is not a key method by which young children recall lists of pictures.     

Differential effects of cue specificity and list length on the prospective and retrospective prospective-memory components

Event-based prospective memory (PM) refers to the ability to remember to perform an intended action at an appropriate moment that is indicated by some cue. It has been shown that increasing the number of PM cues as well as decreasing the cues' specificity can impair PM performance. Although both manipulations result in similar detrimental effects to PM accuracy, they may affect different underlying cognitive processes. To test this hypothesis, we manipulated both the number of cues and cue specificity and found the expected detrimental effects on PM accuracy. Analyses with the multinomial model of PM, which considers false PM responses and accounts for guessing biases, imply that the cue-list-length manipulation affected cue singularisation as well as intention retrieval, whereas cue specificity selectively affected intention retrieval. These results are original evidence that the performance decrements from the two manipulations have different cognitive underpinnings.     

Testing the concurrent and predictive relations among articulation accuracy, speech perception, and phoneme awareness

The relations among articulation accuracy, speech perception, and phoneme awareness were examined in a sample of 97 typically developing children ages 48 to 66 months. Of these 97 children, 46 were assessed twice at ages 4 and 5 years. Children completed two tasks for each of the three skills, assessing these abilities for the target phoneme /r/ and the control phoneme /m/ in the word-initial position. Concurrent analyses revealed that phoneme-specific relations existed among articulation, awareness, and perception. Articulation accuracy of /r/ predicted speech perception and phoneme awareness for /r/ after controlling for age, vocabulary, letter-word knowledge, and speech perception or phoneme awareness for the control phoneme /m/. The longitudinal analyses confirmed the pattern of relations. The findings are consistent with a model whereby children's articulation accuracy affects preexisting differences in phonological representations and, consequently, affects how children perceive, discriminate, and manipulate speech sounds.