The Application of Statistical Methods to the Problems of Psychophysics

The Application Of Statistical Methods To The Problems Of Psychophysics by F. M Urban. The book provides a rigorous introduction to applying probability theory and statistical analysis to psychophysical measurements. Urban describes experimental procedures for determining the threshold of difference and generalizes these methods to questions of maximum and minimum stimulation. The work emphasizes that data from psychological experiments are observations of relative frequencies and that their interpretation rests on the concept of a system of causes producing probabilities for different classes of judgments (heavier, lighter, equal). The text details the arrangement of experiments with hollow brass weights, the use of a metronome to regulate lifting, and the recording of judgments of 'heavier', 'lighter', or 'equal', with varying degrees of confidence. It introduces the notion of the probabilities of judgments and derives the coefficient of divergence and the component of physical variation to assess dispersion across series. The method of just perceptible differences is developed, including the regularization through several arrangements of stimuli, interpolation considerations, and the relation to error methods. Urban then surveys psychometric functions, equality judgments, and broader issues in the philosophy of scientific measurement, culminating in a general discussion of probabilistic descriptions of mental life.