A nonmetric variety of linear factor analysis

The numbers in each column of ann ×m matrix of multivariate data are interpreted as giving the measured values of alln of the objects studied on one ofm different variables. Except for random error, the rank order of the numbers in such a column is assumed to be determined by a linear rule of combination of latent quantities characterizing each row object with respect to a small number of underlying factors. An approximation to the linear structure assumed to underlie the ordinal properties of the data is obtained by iterative adjustment to minimize an index of over-all departure from monotonicity. The method is “nonmetric” in that the obtained structure in invariant under monotone transformations of the data within each column. Except in certain degenerate cases, the structure is nevertheless determined essentially up to an affine transformation. Tests show (a) that, when the assumed monotone relationships are strictly linear, the recovered structure tends closely to approximate that obtained by standard (metric) factor analysis but (b) that, when these relationships are severely nonlinear, the nonmetric method avoids the inherent tendency of the metric method to yield additional, spurious factors. From the practical standpoint, however, the usefulness of the nonmetric method is limited by its greater computational cost, vulnerability to degeneracy, and sensitivity to error variance.     

A large computer looks at the distributed computing problem

The various ways in which a large computer can be of service to a small laboratory processor are discussed. The Cambridge Project and the Consistent System (a collection of programs, models, and data) are briefly described.     

Pavlovian theory of dreaming

Pavlovian theory provides a conceptual scheme for the examination of recent empirical data on dreaming. Consistencies are examined between these data and Pavlovian theory of the general properties of the nervous system. An equation is made between Pavlov’s excitation-inhibition continuum, the REM-NREM cycle, and the basic rest-activity cycle. Speculations, based on this equation, are made about the inter-relation between waking and sleeping behaviors. Testable hypotheses are suggested throughout the paper.     

Learning to utilize information presented over two sensory channels

In three separate experiments, Ss were provided with auditory, visual, or simultaneous auditory and visual information in a classification task. Difficulty of classification was manipulated by varying the stimulus exposure duration. Consistent bisensory facilitation effects were noted for later trials, with interference evident on earlier trials. Exposure duration influenced rate and not amount of learning, with bisensory performance being most affected by duration. A transfer paradigm was used in Experiment III, and little if any transfer was noted between unisensory and bisensory stimulus conditions. It was concluded that Ss were extracting the most salient bisensory stimulus components from the auditory and visual modes of information into a unidimensional information configuration.     

A random-pulse generator with adjustable pulse duration

An electronic device for generating random pulses of adjustable duration within a minimal and maximal interval boundary is described. The instrument is useful in electro- and psychophysiological research where randomization of stimulus input is required.     

Crossbreeding and special handling of genetically nervous dogs

In studies of a healthy (A) and a nervous (E) genetic strain of pointer dogs, two techniques were utilized to determine environmental and hereditary influences. These were (1) reciprocal crossbreeding to females from each strain, and (2) split litters with special handling and human attention to one-half of each litter of nervous strain origin from birth through age six months. It is not possible to attribute the nervous behavior to an environmental mother effect since crossbred AE and EA offspring were highly similar on behavior tests of brief exploratory activity, rigid posturing to a loud noise and human avoidance behavior. Furthermore, special handling had almost no success in normalizing the responses of the nervous dogs in the behavior tests or in heart rate response to Effect-of-Person procedures. The results show that the biologic origin of the dogs is the most likely source of the nervousness; this has placed severe limits thus far on attempts to attenuate the condition.