W. H. Ittelson, The Ames Demonstrations in Perception (Princeton University Press, Princeton, New Jersey, 1952), pp. 50–52.

R. K. Luneburg, Mathematical Analysis of Binocular Vision (Princeton University Press, Princeton, New Jersey, 1947). (a) Pp. 89–94.

R. K. Luneburg, “Metric Methods in Binocular Visual Perception” in Studies and Essays, Courant Anniversary Volume (Interscience Publishers, Inc., New York, 1948).

Luneburg used the phrase“visual space” to refer to the three-dimensional space in which are localized the colors, brightnesses, and objects of which we are immediately aware (Ref. 1, p. 1). This phenomenological definition, however, is not essential to his development, since only the mathematical properties of the space enter into his argument. Blank (Ref. 13, p. 329) refers to the visual space as a class of elements called points, having among them two relations designated as length ordering and alignment. These relations are defined mathematically only in terms of the axioms. In any experiment a finite number of these points are interpreted as points of the stimulus and the relations are interpreted as the relations among the stimulus points after they have been arranged to satisfy some instructional criterion.

H. von Helmholtz and J. Southhall (translator), Physiological Optics (Optical Society of America, Rochester, New York, 1925), Vol. 3, p. 318.

H. Busemann, Metric Methods in Finsler Spaces and in the Foundations of Geometry (Princeton University Press, Princeton, New Jersey, 1942), p. 115.

H. Busemann, The Geometry of Geodesies (Academic Press Inc., New York, 1955).

C. J. Campbell, “An Experimental Investigation of the Size Constancy Phenomenon” (unpublished dissertation, Columbia University, 1952).

Prior to this study a pilot study was carried out. It differed from the present study in that it was done with a slightly less precise apparatus, involved 20 rather than 24 settings of each segment, and had some minor differences in procedure. Two observers, 01 and 02, were used; 02 was the same observer as 02 of the present study. The mean shortest join length was found to be 0.002 cm for 01 and 0.067 cm for 02; the first of these was not significantly different from 0 at the 0.05 level of significance; the second was significantly different.

Since significantly large correlations exist among the components of the shortest join, the three tests for each observer are not independent. This weakness could be overcome by calculating the sum of the three components for each setting and testing the mean of these sums against an estimate of its standard deviation. However, since in all cases the D component is considerably larger than the other two, it is evident that the result of such a test would be essentially the same as the result given in Table II for the D component.