The general psychophysical differential equation: A comparison of three specifications

The general psychophysical differential equation, dy/dx = W₂(y)/W₁(x), with the solution y = f(x), where x and y are subjective variables and W₁ and W₂ their subjective Weber functions, is (a) compared with a corresponding functional equation, and (b) studied from a stochastic point of view by error calculus, Methods for evaluating and handling divergences are proposed and illustrated for a number of combinations of Weber functions. It is shown that either the differential: and the functional equations have the same solution or the difference between the solutions is negligible compared to empirical scatter. The error calculus gives the same result: either no error at all or a negligible one.     

“Motority” in three functional domains of word meaning

This article deals with relations between mental motor evoking capacity (motority, Putnoky, 1975a, b) and the sensory and affective connotations, as well as certain denotations, of words. Sensory connotation is indexed by degree of visual auditory, and tactual imagery, as assessed on a 7-point scale, while affective connotation is indexed by the usual three factors tapped by the semantic differential. The aspect of denotation used here is the abstractness value of words, also based on rating by scales. The general conclusions about motority are as follows: (1) the finding that its sensory connotations are mainly via theauditory modality of imagery, and especially in correlation with abstract words, suggests that it has a specific function that may have originated in the processing of heard language; (2) it correlates with thepotency andactivity factors of affective connotation, which suggests the existence of a nonspecific function or component in it; (3) correlations with abstractness reflect that motority must have a denotative function or component, especially in case of abstract verbal stimuli. On the basis of the results, motority is supposed to play a regulative role at higher levels of the organization of word meaning in particular.Computation of SD data was carried out under the direction and according to the programming of Professor William H. May in the Center for Comparative Psycholinguistics (Director: Charles E. Osgood), University of Illinois, Urbana-Champaign, with support provided by the National Institute of Mental Health (NIMH 07705) and National Science Foundation (NSF GS 360 and 2012-X).     

Two-step weighted least squares factor analysis of dichotomized variables

A two-step weighted least squares estimator for multiple factor analysis of dichotomized variables is discussed. The estimator is based on the first and second order joint probabilities. Asymptotic standard errors and a model test are obtained by applying the Jackknife procedure.     

Time errors are perceptual

Summary Methods for the measurement of time-errors (TEs) in the comparison of successive stimulus magnitudes are discussed. Combining a Thurstonian scaling method with the assumption of a fixed subjective width of the equal category, independent of stimulus level, a ratio scale for subjective differences within pairs of successive stimuli is derived. In a tone duration comparison experiment, with the TE defined in the terms of these subjective duration differences, data from four experimental groups were compared, the groups using different modes of judging and responding. Only minor effects of this factor were found, and hence it is concluded that the TE is a true perceptual phenomenon rather than an effect of response bias, criterion bias, or mediating verbal responses to the absolute level of stimulation. The quantitative results are interpreted in terms of a general model for the comparison of successive stimuli, employing the concepts of adaptation and differential weighting of sensation magnitudes.This investigation was supported by grants to the author from the University of Stockholm and from the Swedish Office of Administrative Rationalization and Economy (for computer time), and by grants to Mats Björkman and Hannes Eisler from the Swedish Council for Social Science Research, whose free consultation service at the Department of Statistics, University of Stockholm, also benefited the author.