A comparison of four methods of constructing factor scores

Four least-squares methods for constructing factor scores have been described in the literature. The formal properties of these scores are developed, and they are compared in terms of four generally desirable properties of constructed factor scores. In particular, it is shown that two of the methods yield scores that are conditionally unbiassed, and univocal in the sense of Guilford and Michael, though not orthogonal, while one of the other methods yields orthogonal scores.It is shown that constructed factor scores cannot be simultaneously univocal and orthogonal, unless we choose the special basis in factor space given by Canonical Factor Analysis.The general problem of choosing between the methods is discussed, on the basis of the theoretical relations obtained.EDITOR'S NOTE: The reader will quickly discover that this article develops several of the generalizations given in the second part of the preceding article, On Factors and Factor Scores. Independent development of the same generalizations is, of course, not a new phenomenon. Because the Presidential Address automatically is accepted for publication and given space in the December issue, it was decided that the only fair thing to do was to print this article in the same issue.     

The functional architecture of instrumental conditional reflexes

The author holds the opinion that the so-called operant, or instrumental, conditional reflexes, in spite of differences in specific features, do not differ in principle from classical conditional reflexes, and therefore they should not be opposed to them, or be placed in a special or separate group. In support of this opinion, this paper presents data which show that:

1. 1.
   In the establishment of classical and operant conditional reflexes, two-way conditional connections—direct and reverse—are formed.     

A portable solid-state stimulus programmer

This paper describes a miniature solid-state timing device particularly useful for presenting repetitive signal sequences in classical conditioning experiments. The device uses transistors, silicon-controlled rectifiers, and four-layer diodes, and is powered by 20-hour rechargeable nickel-cadmium batteries. Visual monitoring is provided for the unelapsed ITI and battery charge. Ranges are; ITI, 15 sec. to 5 min., ISI, 0–50 secs., CS duration, 0.5 to 30 sec., US duration, 0.5 to 30 sec. The CS and US durations are adjustable and the onset of the latter may be at any time after the CS onset. Cycling is continuous and automatic; the outputs are in the form of relay closures.     

Detection of auditory signals presented at random times

One hundred msec tones of 1000 Hz at four intensities were presented according to two Poisson schedules in a background of wide band noise and as increments to a 1000 Hz tone. Each 15 minute test session was run under fixed conditions and one of three instructions to detect the signals; the two Ss were free to respond at any time. The data analyzed were several inter-response and signal-response distributions, and they were compared with a theory due to Luce. Discrepancies between the theory and data indicate needed modifications of both the theory and the procedure.     

The effect of angle between the oblique lines on the decrement of the Müller-Lyer illusion with extended practice

Early work on the Müller-Lyer illusion had indicated that it disappears with extended practice. The present experiment failed to confirm this finding. The magnitude of the illusion decreased for approximately 500 trials, but showed no further change over an additional 500 trials. The rate of the practice decrement was inversely related to the size of the angle formed by the oblique lines of the figure.