SPA TIAL PERCEPTION. 1135 



(Eaehlmann and Witkowski). 1 Under certain conditions, as in stereoscopic 

 combination, we localise quite independently of the degree of convergence. 



Accommodation and localisation in depth. Several investiga- 

 tions have been carried out to determine whether sensations of move- 

 ment may act as subsidiary factors in localisation. It is impossible 

 to test the influence of convergence experimentally, since the binocular 

 retinal mechanism cannot be excluded when both eyes are used. In 

 experiments with one eye the psychological factors must be absolutely 

 excluded. In order to do this most completely, Hillebrand, 2 following a 

 suggestion of Hering, has used the dividing line between two surfaces. 

 The observer looks through an aperture at a fixed white screen, one-half 

 of which is hidden by a black screen with sharply cut edge which is 

 capable of being moved towards or away from the eye. It is found that 

 the observer is totally unable to tell the distance of the black screen, or 

 whether it is moving towards or away from him, or even whether it is 

 moving at all. Wundt 3 found that he was able to detect the approach 

 of a vertically suspended thread when there was no appreciable change 

 in its apparent size, and supposed that this was due to the influence of 

 accommodation. He was only able to detect recession of the thread 

 from the eye when it had undergone a visible change in size. Arrer 4 

 has also obtained results with the the thread experiment, which he thinks 

 can only be explained by the influence of movement factors. Hillebrand 

 found when his screen was exposed at intervals at different distances, 

 that some individuals were able to detect whether the screen had been 

 moved towards or away from the eye during the interval, and this has 

 been confirmed by Dixon 5 and Arrer. It is not easy to feel quite certain 

 that psychological factors have been absolutely excluded, and the evidence 

 shows that if sensations arising from accommodation are of any influence 

 at all in localisation, this influence is very slight and limited to certain 

 individuals. 



Hillebrand argued that, owing to the close association between con- 

 vergence and accommodation, his experiment disproving the influence of 

 accommodation also disproved the influence of convergence in depth 

 perception, but the association is not close enough to justify this 

 argument, and the possibility still remains that convergence may have 

 some influence, although there is no positive evidence in favour of it. 



The cyclopean eye. Some of the chief investigators of binocular 

 vision (Helmholtz, Hering, and less definitely Bonders) have adopted 

 the idea that we localise objects in space in reference to the middle 

 point between the two eyes, and as if seen by an eye in that position. 

 If two parallel lines are drawn at a distance apart, corresponding 

 to the interocular distance, and are observed so that one line lies 

 in the line of vision of each eye, the combined image of the two lines 

 will be seen in the median plane between the two eyes, as if it were 

 coming from a middle eye. The same appearance is produced with 

 converging lines, provided that the lines of vision converge to a corre- 

 sponding degree. 6 This tendency is so strong, that, in monocular vision, 



1 Arch.f. PhysioL, Leipzig, 1877, S. 463. 



' 2 Ztschr.f. Psychol. u. PhysioL d. Sinnesorg., Hamburg u. Leipzig, 1894, Bd. vii. S. 97. 



3 "Beitr. z. Theorie d. Sinneswahrnahrung," 1862, S. 105. 



4 Phil. Stud., Leipzig, 1896-1897, Bd. xiii. S. 116 and 222. 



5 Mind, London and Edinburgh, 1895, vol. iv. p. 195. 



6 For other experiments illustrating the cyclopean eye, see Towne, Guy's ffosp. Rep., 

 London, 1865, vol. xi. p. 144 ; and Le Conte, "Sight," 1883, p. 213. 



