The Measurement of Discriminability

  title={The Measurement of Discriminability},
  author={Edward R. Grossman},
  journal={Quarterly Journal of Experimental Psychology},
  pages={176 - 195}
  • E. R. Grossman
  • Published 1 April 1955
  • Computer Science
  • Quarterly Journal of Experimental Psychology
Abstract The need for a quantitative measure of “Discriminability” is pointed out. A formula is required which will enable one, knowing the physical measurements of a set of sensory signals, to predict the time taken by a human observer to recognize any one. Existing formulae relate to the two-choice threshold task and to very “easy” multi-choice ones. A more general treatment is attempted in order to cover the region between these two special cases. For the limited case of two-choice one… 

Perceptual Indices of Performance: The Measurement of ‘Inspection Time’ and ‘Noise’ in the Visual System

Analysis of the results of the three experiments carried out, it is suggested that measures of inspection time and noise, together with a third index related to the degree of caution exercised by an observer, appear to be stable and consistent descriptors of performance.

Test of the Accuracy of Crossman's Confusion-Function

  • W. McCoy
  • Psychology
    Perceptual and motor skills
  • 1963
The data analysis provided striking confirmation of Crossman's hypothesis of linearity and added evidence to the possibility that the Confusion-Function might apply to discrimination in general.

Information Load and Discriminability

The two experiments summarized here show interactions between size of stimuli, illumination-level, and the number of items between which discriminations are made, and how long it takes subjects to differentiate between two stimuli.

Evidence for an accumulator model of psychophysical discrimination.

It is suggested that an ‘ accumulator ’ model, in which sampled events may vary in magnitude as well as probability, can be developed to give a good account of much of the available data on psychophysical discrimination.

What is the basis of choice reaction-time?

The inequalities observed are incompatible with several types of model which have been proposed to account for the linear rise of reaction-time with the logarithm of degree of choice; in particular, the serial classification model which assumes a simple progressive dichotomization, and models which assume that evidence from all the possible signal sources is examined simultaneously.

A Mathematical Model for Perception and a Theoretical Confusion Function

A mathematical model of perception processes was outlined, from which a theoretical confusion function was derived that depended upon the ratio of the stimulus values concerned, and not upon their absolute magnitudes.

On the modeling of the performances of the human brain in a two-choice task involving decoding and memorization of simple visual patterns

It is shown that the experimental design allows a quantitative evaluation of those factors which are relevant to the decoding of patterns of different complexities, and the necessity for a nonlinear summation of the decoded information and its decay with time.


Three models to account for discrimination times are examined : one being derived from information-theory, another being Grossman's Confusion Function and the third a form of Sequential Sampling

Matching performance and the similarity structure of the stimulus set.

  • W. B. Crist
  • Psychology
    Journal of experimental psychology. General
  • 1981
Lockhead's holistic-discriminability model provides a framework for interpreting demonstrations that matching performance does not depend only on the stimuli physically present, and on the similarity or discriminability between the presented stimuli and other stimuli the observers know might be presented.

Grossman's Confusion-function and Multi-choice Discrimination

In a recent paper Laming (1962) discusses afresh the relation between multi-choice reaction times and both the difficulty of discrimination and the entropy of the choice situation, a topic first



On the Rate of Gain of Information

The principal finding is that the rate of gain of information is, on the average, constant with respect to time, within the duration of one perceptual-motor act, and has a value of the order of five “bits” per second.

Entropy and choice time: The effect of frequency unbalance on choice-response

A human subject making a sequence of choice-responses is considered as a channel transmitting information and entropy is reduced by unbalance in the relative frequency of the possible signals according to the formula, sum (p log p).

Stimulus information as a determinant of reaction time.

  • R. Hyman
  • Psychology
    Journal of experimental psychology
  • 1953
In the typical reaction-time experiment, S's reaction time is greater when he has to respond differentially to one of two equally probable stimuli instead of to just one stimulus, but this becomes even more significant when looked at from the standpoint of modern communication theory.

Elewente Der Psycho-physih 11

  • Leipzig. HELSON,
  • 1860

Adaptation-level as a basis for a quantitative theory of frames of HEXMON

  • 1906

Planned Seeing

  • H.M.S.O., London.
  • 1950