Some comments in connection with Rozeboom's linear correlation theory

It is shown that certain correlation measures suggested by information theory are simple functions of the determinants of the correlation matrices involved, if multivariate normality is assumed. This illuminates and in a sense strengthens Rozeboom's point in [3], that information theoretical statistics have classical correlational counterparts.     

Une hypothèse sur l'extension des relations finies et sa vérification dans certaines classes particulières

Sans résumé     

Problems of Work Evaluation

In reviewing the history of the development of techniques of work evaluation, four different approaches are appraised: (1) the mental testing approach, (2) the job analysis approach, (3) the work-sample approach, and (4) the situational assessment approach. Each of these sets of methods is found to arise from somewhat different sectors of society, to serve differing interests, and to have differing objectives. After examination of the merits and limitations of each of the four approaches, the conclusion is drawn that “no one of these four methods can do everything; each deals with a restricted facet of a many-faceted problem.” Attention is called to the general need for validation through professional observation in unprotected industrial settings.     

Saturation of red: A prothetic continuum

A battery of procedures was used to scale the saturation of colors produced by mixtures of red and gray papers. By direct magnitude estimation, the apparent saturation was found to grow as the1.7 power of the percentage of red in the mixture. The power law was confirmed by the cross-modality matching of loudness to saturation. The inverse continuum, paleness, was also scaled. Saturation appears to be a prothetic continuum because the category scale is curved and the subjective size of the jnd increases as saturation increases.     

Regression effect in psychophysical judgment

Psychophysical judgment, like all other kinds of judgment, involves a matching or equating of two different domains. When the judgment involves the matching of values on two perceptual continua, the observer tends, on the average, to constrict the range of his adjustments on whichever variable is placed under his control. When the observer adjusts each variable in turn, two different regression lines are produced. This regression effect presumably occurs whenever the results of the matching judgments yield less than a perfect correlation. Illustrative examples are given for the continua, loudness, vibration, brightness, and duration.     

Binocular summation in the perception of form at brief durations

Visual form identification at brief durations was studied under: (a) monocular presentation; (b) dichopic presentation where the same form was presented successively on noncorresponding areas; and (c) dichopic presentation where the same form was presented on corresponding areas simultaneously and successively. Form identification for noncorresponding area dichopic presentation was at the level to be expected from 2independent chances to perceive. Both simultaneous and successive dichopic presentation on corresponding areas gave identification accuracy significantly above the level predicted by the assumption of independence. However, the binocular summation was not complete. When the same amount of energy entering the visual system in a binocular presentation was given in a monocular stimulation, the latter condition gave significantly better identification.