The estimation of ultrametric and path length trees from rectangular proximity data

A least-squares algorithm for fitting ultrametric and path length or additive trees to two-way, two-mode proximity data is presented. The algorithm utilizes a penalty function to enforce the ultrametric inequality generalized for asymmetric, and generally rectangular (rather than square) proximity matrices in estimating an ultrametric tree. This stage is used in an alternating least-squares fashion with closed-form formulas for estimating path length constants for deriving path length trees. The algorithm is evaluated via two Monte Carlo studies. Examples of fitting ultrametric and path length trees are presented.G. De Soete is Aspirant of the Belgian Nationaal Fonds voor Wetenschappelijk Onderzoek at the University of Ghent, Belgium. W. S. DeSarbo and J. D. Carroll are Members of Technical Staff at AT&T Bell Laboratories, Murray Hill, N.J. G. W. Furnas is Member of Technical Staff at Bell Communications Research, Murray Hill, N.J.     

The similarity and diversity of semantic relations

There is a rich variety of semantic relations in natural languages. Subjects’ perceptions of similarities among relations were studied for a wider variety of relations than had been used in previous studies. Forty subjects sorted 31 cards bearing five example pairs of each of 31 semantic relations. Subjects were able both to distinguish the relations and to perceive their similarities. A hierarchical clustering analysis of the sorting data indicated that the subjects perceived five families of semantic relations (contrasts, class inclusion, similars, case relations, and part-wholes). The five families were distinguished in terms of three properties of semantic relations: contrasting/noncontrasting, logical/pragmatic, and inclusion/noninclusion. Within each family, relations also were sorted in ways consistent with their defining properties. Relations were therefore viewed not as unanalyzable primitives, but in terms of the relational properties that distinguished them.     

Confidence regions for INDSCAL using the jackknife and bootstrap techniques

Bootstrap and jackknife techniques are used to estimate ellipsoidal confidence regions of group stimulus points derived from INDSCAL. The validity of these estimates is assessed through Monte Carlo analysis. Asymptotic estimates of confidence regions based on a MULTISCALE solution are also evaluated. Our findings suggest that the bootstrap and jackknife techniques may be used to provide statements regarding the accuracy of the relative locations of points in space. Our findings also suggest that MULTISCALE asymptotic estimates of confidence regions based on small samples provide an optimistic view of the actual statistical reliability of the solution. The authors wish to thank Geert DeSoete, Richard A. Harshman, William Heiser, Jon Kettenring, Joseph B. Kruskal, Jacqueline Meulman, James O. Ramsay, John W. Tukey, Paul A. Tukey, and Mike Wish. Sharon L. Weinberg is a consultant at AT&T Bell Laboratories, Murray Hill, New Jersey 07974.     

Pavlov’s position on the effects of decortication on conditional reflexes

This paper examined D. Joravsky’s (1989) hypothesis that I.P. Pavlov dogmatically refused to acknowledge that classical conditioning can be mediated by subcortical regions of the large cerebral hemispheres. Decortication literature from 1901 to 1936 was reviewed. The early studies available to Pavlov, who died in 1936, showed that decortication does not allow the establishment of new or retaining of old conditional reflexes (CRs). G.P. Zelenyi’s later experiments(1930) suggested that the establishment of primitive CRs in decorticated dogs was possible. Pavlov never denied this possibility but cautioned that Zelenyi’s experiments could have been methodologically flawed. Although Joravsky’s original hypothesis on Pavlov’s position on the relation between decortication and the establishment of CRs is by and large accepted, it must be stressed that Pavlov’s theory of higher nervous activity was primarily concerned with the function of the brain in the higher organism’s struggle for existence. Within this context the cortical, rather than subcortical, processes play the decisive role in the organism’s adaptation to the changing external environment.     

CHANGES IN STUDENT AND TEACHER RESPONSES IN OBSERVED AND GENERALIZED SETTINGS AS A FUNCTION OF SUPERVISOR OBSERVATIONS

Two studies are reported in which the effects of supervisor observations of teachers' performance rate and accuracy on both teachers' and severely handicapped students' behaviors were compared with baseline supervisor observations that did not specify rate and accuracy feedback. The latter observation procedure (nonspecific feedback) was more typical of the existing practices of school supervisors. The dependent variables for the teachers were rate and accuracy of teacher behaviors to student responses in individualized instructional settings involving discrete trials or task-analysis steps. The dependent variables for students were the rates of correct and incorrect responding to teacher presentations during the supervisor observation periods. Both studies used a multiple baseline design. The results of Study 1 showed that there were educationally significant changes in teacher and, in turn, student performance as a result of the use of the rate and accuracy procedure. The results of Study 2 replicated those of Study 1 while correcting limitations of Study 1. This latter study also demonstrated that the observation effects of the procedure generalized to teacher performance throughout the day. The results are discussed in terms of the efficiency and cost effectiveness of the procedure to train, monitor, and reinforce teacher effectiveness.     

Decision-making style and vocational maturity

This study examines the relationship between decision-making style, scholastic achievement, and vocational maturity for college students. A positive relationship between a rational decision-making style and the multifaceted construct of vocational maturity has been inherent in much theory and has provided the basis for most practice. The hypothesized relationship between rationality and attitudinal and cognitive maturity was not supported by the results reported here. Only a combination of the scholastic achievement variable and a lack of dependent decision style was found to be even moderately predictive of vocational maturity. Implications for practice and future research are discussed.     

Spatial versus tree representations of proximity data

In this paper we investigated two of the most common representations of proximities, two-dimensional euclidean planes and additive trees. Our purpose was to develop guidelines for comparing these representations, and to discover properties that could help diagnose which representation is more appropriate for a given set of data. In a simulation study, artificial data generated either by a plane or by a tree were scaled using procedures for fitting either a plane (KYST) or a tree (ADDTREE). As expected, the appropriate model fit the data better than the inappropriate model for all noise levels. Furthermore, the two models were roughly comparable: for all noise levels, KYST accounted for plane data about as well as ADDTREE accounted for tree data. Two properties of the data proved useful in distinguishing between the models: the skewness of the distribution of distances, and the proportion of elongated triangles, which measures departures from the ultrametric inequality, Applications of KYST and ADDTREE to some twenty sets of real data, collected by other investigators, showed that most of these data could be classified clearly as favoring either a tree or a two-dimensional representation.A portable PASCAL program implementing the Sattath and Tversky [1977] ADDTREE algorithm is available from J. Corter, Department of Psychology, Stanford University, Stanford, California 94305.     

Using the FORTH language in real-time computer applications

The FORTH computer language is described, with emphasis on aspects relevant to its use in the real-time control of psychological experiments via microcomputers. Among FORTH’s advantages are speed, compactness, and flexibility. Potential disadvantages include unusual syntax and lack of file-oriented disk access. Two sample FORTH-based systems are described.