Statistical inference based on latent ability estimates

The quality of approximations to first and second order moments (e.g., statistics like means, variances, regression coefficients) based on latent ability estimates is being discussed. The ability estimates are obtained using either the Rasch, or the two-parameter logistic model. Straightforward use of such statistics to make inferences with respect to true latent ability is not recommended, unless we account for the fact that the basic quantities are estimates. In this paper true score theory is used to account for the latter; the counterpart of observed/true score being estimated/true latent ability. It is shown that statistics based on the true score theory are virtually unbiased if the number of items presented to each examinee is larger than fifteen. Three types of estimators are compared: maximum likelihood, weighted maximum likelihood, and Bayes modal. Furthermore, the (dis)advantages of the true score method and direct modeling of latent ability is discussed.     

“Minimax length links” of a dissimilarity matrix and minimum spanning trees

A theorem is proved stating that the set of all minimax links, defined as links minimizing, over paths, the maximum length of links in any path connecting a pair of objects comprising nodes in an undirected weighted graph, comprise the union of all minimum spanning trees of that graph. This theorem is related to methods of fitting network models to dissimilarity data, particularly a method called Pathfinder due to Schvaneveldt and his colleagues, as well as to single linkage clustering, and results concerning the relationship between minimum spanning trees and single linkage hierarchical trees.Acknowledgments: The author thanks Phipps Arabie, Lawrence J. Hubert, and K. Christoph Klauer for a number of helpful suggestions and comments on various aspects of this paper.     

Choice and foraging: the effects of accessibility on acceptability.

Pigeons responded in a successive-encounters choice procedure in which accessibility of the less profitable of two outcomes varied either in terms of probability of encounter or search time to encounter (keeping search time to the more profitable outcome constant). When the less profitable outcome was made more probable its acceptance became more likely. However, when search time to encounter the less profitable outcome was shortened, its acceptance became less likely. Both results are consistent with the delay-reduction hypothesis and with an optimality model developed for application to the successive-encounters choice procedure.     

Comparing item characteristic curves

Test items are often evaluated and compared by contrasting the shapes of their item characteristics curves (ICC's) or surfaces. The current paper develops and applies three general (i.e., nonparametric) comparisons of the shapes of two item characteristic surfaces: (i) proportional latent odds, (ii) uniform relative difficulty, and (iii) item sensitivity. Two items may be compared in these ways while making no assumption about the shapes of item characteristic surfaces for other items, and no assumption about the dimensionality of the latent variable. Also studied is a method for comparing the relative shapes of two item characteristic curves in two examinee populations.The author is grateful to Paul Holland, Robert Mislevy, Tue Tjur, Rebecca Zwick, the editor and reviewers for valuable comments on the subject of this paper, to Mari A. Pearlman for advice on the pairing of items in the examples, and to Dorothy Thayer for assistance with computing.     

Maximum likelihood additivity analysis

A maximum likelihood estimation procedure was developed to fit unweighted and weighted additive models to conjoint data obtained by the categorical rating, the pair comparison or the directional ranking method. The scoring algorithm used to fit the models was found to be both reliable and efficient, and the program MAXADD is capable of handling up to 300 parameters to be estimated. Practical uses of the procedure are reported to demonstrate various advantages of the procedure as a statistical method.The research reported here was supported by Grant A6394 to the author from the Natural Sciences and Engineering Research Council of Canada. Portions of this research were presented at the Psychometric Society meeting in Iowa City, Iowa, in May, 1980.Thanks are due to Jim Ramsay, Justine Sergent and anonymous reviewers for their helpful comments.Two MAXADD programs which perform the computations discussed in this paper may be obtained from the author.     

The role of contingencies and "principles of behavioral variation" in pigeons' pecking

Staddon and Simmelhag's proposal that behavior is produced by “principles of behavioral variation” instead of contingencies of reinforcement was tested in two experiments. In the first experiment pigeons were exposed to either a fixed-interval schedule of response-contingent reinforcement, an autoshaping schedule of stimulus-contingent reinforcement, or a fixed-time schedule of noncontingent reinforcement. Pigeons exposed to contingent reinforcement came to peck more rapidly than those exposed to noncontingent reinforcement. Staddon and Simmelhag's “principles of behavioral variation” included the proposal that patterns (interim and terminal) were a function of momentary probability of reinforcement. In the second experiment pigeons were exposed to either a fixed-time or a random-time schedule of noncontingent reinforcement. Pecking showed a constant frequency of occurrence over postfood time on the random-time schedule. Most behavior showed patterns on the fixed-time schedule that differed in overall shape (i.e., interim versus terminal) from those shown on the random-time schedule. It was concluded that both the momentary probability of reinforcement and postfood time can affect patterning.     

基于语料库统计的“音-形”激活概率及加工机制

为了了解心理词典中语音与汉字的联结情况以及从音到形的加工机制,研究首先计算出了基于语料库统计的“音-形”激活概率(某个音节激活其对应汉字的概率),并通过一个听写实验来验证这种激活概率是否反映实际加工中的激活情况。结果证明基于语料统计得到的“音-形”激活概率基本可以预测实际的“音-形”激活情况,表现为:“音-形”激活概率越高的汉字被激活的概率和强度越高,并且有明显的优势字效应; 该统计概率具有可独立于其他影响因素的预测功能。     

Prime and probability: causal knowledge affects inferential and predictive effects on self-agency experiences

Experiences of having caused a certain outcome may arise from motor predictions based on action–outcome probabilities and causal inferences based on pre-activated outcome representations. However, when and how both indicators combine to affect such self-agency experiences is still unclear. Based on previous research on prediction and inference effects on self-agency, we propose that their (combined) contribution crucially depends on whether people have knowledge about the causal relation between actions and outcomes that is relevant to subsequent self-agency experiences. Therefore, we manipulated causal knowledge that was either relevant or irrelevant by varying the probability of co-occurrence (50% or 80%) of specific actions and outcomes. Afterwards, we measured self-agency experiences in an action–outcome task where outcomes were primed or not. Results showed that motor prediction only affected self-agency when relevant actions and outcomes were learned to be causally related. Interestingly, however, inference effects also occurred when no relevant causal knowledge was acquired.     

Effects of reinforcer probability, delay, and response requirements on the choices of rats and pigeons: possible species differences

In Experiment 1 with rats, a left lever press led to a 5-s delay and then a possible reinforcer. A right lever press led to an adjusting delay and then a certain reinforcer. This delay was adjusted over trials to estimate an indifference point, or a delay at which the two alternatives were chosen about equally often. Indifference points increased as the probability of reinforcement for the left lever decreased. In some conditions with a 20% chance of food, a light above the left lever was lit during the 5-s delay on all trials, but in other conditions, the light was only lit on those trials that ended with food. Unlike previous results with pigeons, the presence or absence of the delay light on no-food trials had no effect on the rats' indifference points. In other conditions, the rats showed less preference for the 20% alternative when the time between trials was longer. In Experiment 2 with rats, fixed-interval schedules were used instead of simple delays, and the presence or absence of the fixed-interval requirement on no-food trials had no effect on the indifference points. In Experiment 3 with rats and Experiment 4 with pigeons, the animals chose between a fixed-ratio 8 schedule that led to food on 33% of the trials and an adjusting-ratio schedule with food on 100% of the trials. Surprisingly, the rats showed less preference for the 33% alternative in conditions in which the ratio requirement was omitted on no-food trials. For the pigeons, the presence or absence of the ratio requirement on no-food trials had little effect. The results suggest that there may be differences between rats and pigeons in how they respond in choice situations involving delayed and probabilistic reinforcers.     

Conditional Probability and Defeasible Inference

We offer a probabilistic model of rational consequence relations (Lehmann and Magidor, 1990) by appealing to the extension of the classical Ramsey–Adams test proposed by Vann McGee in (McGee, 1994). Previous and influential models of non-monotonic consequence relations have been produced in terms of the dynamics of expectations (Gärdenfors and Makinson, 1994; Gärdenfors, 1993).Expectation is a term of art in these models, which should not be confused with the notion of expected utility. The expectations of an agent are some form of belief weaker than absolute certainty. Our model offers a modified and extended version of an account of qualitative belief in terms of conditional probability, first presented in (van Fraassen, 1995). We use this model to relate probabilistic and qualitative models of non-monotonic relations in terms of expectations. In doing so we propose a probabilistic model of the notion of expectation. We provide characterization results both for logically finite languages and for logically infinite, but countable, languages. The latter case shows the relevance of the axiom of countable additivity for our probability functions. We show that a rational logic defined over a logically infinite language can only be fully characterized in terms of finitely additive conditional probability. The research of both authors was supported in part by a grant from NSF, and, for Parikh, also by support from the research foundation of CUNY.