A general approach to confirmatory maximum likelihood factor analysis

We describe a general procedure by which any number of parameters of the factor analytic model can be held fixed at any values and the remaining free parameters estimated by the maximum likelihood method. The generality of the approach makes it possible to deal with all kinds of solutions: orthogonal, oblique and various mixtures of these. By choosing the fixed parameters appropriately, factors can be defined to have desired properties and make subsequent rotation unnecessary. The goodness of fit of the maximum likelihood solution under the hypothesis represented by the fixed parameters is tested by a large samplex ² test based on the likelihood ratio technique. A by-product of the procedure is an estimate of the variance-covariance matrix of the estimated parameters. From this, approximate confidence intervals for the parameters can be obtained. Several examples illustrating the usefulness of the procedure are given.This work was supported by a grant (NSF-GB 1985) from the National Science Foundation to Educational Testing Service.     

Discrimination of circles and squares with and without knobs by goldfish

Different groups of goldfish were trained to discriminate between a circle and either a square, a diamond, a square with a knob added at the top or a diamond with a knob. All subjects readily mastered the discrimination, and there was no significant difference between the groups in the number of trials taken to criterion. In transfer tests it was found that: (i) while there was no transfer from square to diamond or vice versa, the presence of knobs on these shapes did mediate transfer from one to the other, provided the knob remained in the top part of the shapes; (ii) all groups learned much more about the tops of the shapes than about the bottom halves; (iii) animals discriminated the square from the circle largely by detecting the presence of a strong horizontal edge at the top of the figure; (iv) two features of the diamond were discriminated--the presence of oblique contours and the existence of a point in the upper half of the shape. The results are discussed in terms of a proposed model of shape recognition.     

Book reviews

Animal Discrimination Learning. Edited by R. M. Gilbert and N. S. Sutherland. London: Academic Press. 1969. Pp. xvi + 501. 1358.

Luce, R. D. and Tukey, J. W. (1964). Simultaneous conjoint measurement: a new type of fundamental measurement. J. math. Psychol. I, 1-27.

Discrimination Learning. By D. A. Riley. Boston: Allyn and Bacon Inc. 1968. Pp. 172. Paperback 21s.

Annual Review of Psychology. Vol. 20. Palo Alto: Annual Reviews Inc. 1969. Pp. ix + 544. $9.00.

Cutaneous Sensation. By David Sinclair. London: Oxford University Press. 1967. Pp. xii + 306. 63s.

Modelling of Thinking and the Mind.. By N. M. Amosov. Translated by L. Finegold. New York: Spartan Books. London: Macmillan. 11967. Pp. xiii + 192.

The Mind of Mnemonist. By A. R. Luria. London: Jonathan Cape. 1969. Pp. xi + 160. 25s.

Determinants of Infant Behaviour IV. Edited by B. M. Foss. London: Methuen. 1969. Pp. 303. £3 10s.

Brain and Behaviour. Volumes 1-4. Edited by Karl Pribram. Harmondsworth: Penguin Books Ltd. 1969. Pp. 496 +573 + 523 + 524. 12s. 6d. per volume.     

Integrated circuit random foreperiod device and digital interval timer1

A digital integrated circuit computer that produces random (as opposed to pseudorandom) foreperiods is described. Also contained in the device is a precision digital interval timer.     

Effects of uncertain warning signals on reaction time

Three experiments were conducted in an attempt to assess the effectiveness of a warning signal in reducing reaction time when (1) the signal to respond (danger signal) follows the warning signal with a probability less than 1.0, and (2) the interval between warning and danger signals (W-D interval) is variable. The required response was the depression of a foot pedal, as in automobile braking. It was determined that probabilistic warning information could be effective if observers made use of the W-D interval to prepare to make the response required by the danger signal. It was noted, however, that observers differed considerably in their tendency to do this. A model was proposed for describing different response strategies.     

Recency discrimination predicted from absolute lag judgments

An experiment is described which tests the plausibility of the assumption that two-alternative forced-choice (2AFC) recency discrimination is based on an independent lag estimate of each item. 39 Ss made absolute recency judgments for items (common nouns) at varying lag intervals while an independent group of 74 Ss made 2AFC judgments for pairs of items at comparable lag combinations. It was found that the absolute judgment data generated accurate predictions of 2AFC performance.     

Sensory scales of taste intensity

The subjective intensity of taste was scaled by the method of magnitude estimation in which Os assigned numbers to designate the apparent strength ofstimulus concentrations. Substances used were sucrose, dextrose, maltose, fructose, saccharin, Sucaryl, sodium chloride, and quinine sulfate. For aqueous solutions of each substance, taste intensity was found to increase as a power function of concentration by weight. Some approximate exponents were: sucrose, 1.3; sodium chloride, 1.4; quinine sulfate. 1.0. The magnitude scale for sucrose was compared with the category scale obtained by a commonly used rating procedure. The category scale turned out to be highly nonlinear.