MATCHING BIAS IN THE SELECTION TASK

A previous study (Evans, 1972) found that subjects tend to match rather than alter named values when constructing verifying and falsifying cases of conditional rules. It was suggested that this tendency (‘matching bias’) might account for the responses normally observed in Wason's (1968, 1969) ‘selection task’. This suggestion was tested by giving subjects the selection task with conditional rules in which the presence and absence of negative components was systematically varied, to see whether subjects consistently attempted to verify the rules (Wason's theory) or whether they continued to choose the matching values despite the presence of negatives, which would reverse the logical meaning of such selections. Significant matching tendencies were observed on four independent measures, and the overall pattern, with matching bias cancelled out, gave no evidence for a verification bias, indicating instead that the logically correct values were most frequently chosen.     

Operationalizing relational turbulence theory: Measurement and construct validation

This study assessed measures operationalizing concepts in relational turbulence theory (Solomon, Knobloch, Theiss, & McLaren, 2016). Studies of relational turbulence in marriage have used measures developed in research on dating couples; therefore, one goal was evaluating these variables in a general population sample of married individuals (N = 1,469). A second goal addressed problems in measures of relational uncertainty by assessing alternative scales. Third, associations between relational turbulence theory variables and measures of cognitive, emotional, and communicative phenomena were estimated. Fourth, the distinctiveness of relational turbulence measures relative to measures of other relational qualities was assessed. The conclusions advance new, Likert relational uncertainty scales and highlight the validity of relational turbulence constructs as distinctive predictors of cognitions, emotions, and communication.     

STATE OF SCIENCE IN INDUSTRIAL AND ORGANIZATIONAL PSYCHOLOGY: A REVIEW OF SELF-REPORTED LIMITATIONS

This study investigated self-reported limitations of published papers as an alternative and novel operationalization of the state of science of industrial and organizational (I-O) psychology. A content analysis was conducted of the reported limitations in every I-O psychology articles published in the Academy of Management Journal, the Journal of Applied Psychology, and Personnel Psychology between 1995 and 2008 (N= 2,402). Articles were coded for the number and types of limitations reported, characteristics of the research design, and topic area. Threats to internal validity were the most often reported limitations. In addition, variations were detected in the reporting of limitations over time, indicating a subtle but steady shift in the focus of I-O psychology research. Implications of these results for the science and practice of I-O psychology and for the use of self-reported limitations in scientific communication are discussed.     

Interpretation and matching bias in a reasoning task

Subjects were required to construct verifying and falsifying cases of conditional rules in which the presence and absence of negative components was varied. Their responses gave some indication of consistent interpretation of the rules, generally conforming to Wason's (1966) idea of a “defective” truth table. Much of their behaviour, however, seemed to be determined by a task variable in the form of a tendency to construct instances which matched, rather than altered, the values named in the rules.     

A theoretical and empirical comparison of the all-or-none multilevel theory and the mixed model

Two theories for stimulus generalization in paired-associate learning (P.A.L.) are compared—the mixed model and the all-or-none multilevel theory. Both theories postulate basic all-or-none learning processes but differ in their mechanism for generalization. The mixed model postulates that generalization occurs only during responding; whereas, the multilevel theory postulates that generalization takes place during learning itself.     

The statistical analysis of general processing tree models with the EM algorithm

Multinomial processing tree models assume that an observed behavior category can arise from one or more processing sequences represented as branches in a tree. These models form a subclass of parametric, multinomial models, and they provide a substantively motivated alternative to loglinear models. We consider the usual case where branch probabilities are products of nonnegative integer powers in the parameters, 0_s1, and their complements, 1 - _s. A version of the EM algorithm is constructed that has very strong properties. First, the E-step and the M-step are both analytic and computationally easy; therefore, a fast PC program can be constructed for obtaining MLEs for large numbers of parameters. Second, a closed form expression for the observed Fisher information matrix is obtained for the entire class. Third, it is proved that the algorithm necessarily converges to a local maximum, and this is a stronger result than for the exponential family as a whole. Fourth, we show how the algorithm can handle quite general hypothesis tests concerning restrictions on the model parameters. Fifth, we extend the algorithm to handle the Read and Cressie power divergence family of goodness-of-fit statistics. The paper includes an example to illustrate some of these results.This research was supported by National Science Foundation Grant BNS-8910552 to William H. Batchelder and David M. Riefer. We are grateful to David Riefer for his useful comments, and to the Institute for Mathematical Behavior Sciences for its support.