SENSORY CAPACITIES AND INTELLIGENCE OF DOGS 



5 



earlier Russian work seems evident. H. 

 M. Johnson has subjected such studies to 

 rigorous criticism. Partly as a result of 

 such criticism, no doubt, radical changes 

 have more recently been made in the 

 technique employed by Pavlov's students. 

 Whether or not these sufficed to entirely 

 eliminate secondary clues cannot be said 

 with our present indirect knowledge of 

 their procedure. They have, apparently, 

 succeeded in eliminating the experimenter 

 from the situation, but this has necessi- 

 tated the introduction of complicated 

 mechanical devices for the presentation of 

 the stimuli, which in themselves may 

 conceivably offer secondary cues. It 

 seems necessary, then, for us to present 

 their results, largely on the basis of their 

 report in Pavlov's recent volume, Con- 

 ditioned Reflexes (3 5), with the hope that 

 it will not be long until their experiments 

 are known to us in greater detail and 

 furthermore, have been verified in other 

 laboratories. 



Concerning the discrimination by the 

 dog of intensity and luminosity differences 

 the Russian work supports the results of 

 Stone (45) mentioned above, who found 

 greater acuity in this respect than any 

 other investigator excepting the Russians. 

 Frolov (35) using the Zimmermann series 

 of grey papers, ranging from white to 

 black, found the dog studied to be slightly 

 superior to human beings, whereas Stone, 

 using a different breed of dog, found them 

 only slightly inferior to man. With 

 respect to the discrimination of form and 

 size, Orbeli reports results conflicting 

 radically with those of other workers, 

 since they seem to indicate discrimination 

 of a large letter T from such forms as a 

 square, a circle and so on. The luminos- 

 ity and size factors, however, were appar- 

 ently uncontrolled. It seems that the 

 work on size discrimination must be 

 thrown out on the same grounds. Fur- 



thermore, Orbeli's work was done in 1907 

 before the improved technique was 

 introduced (57) in Leningrad. The work 

 of Shenger-Krestovnikova (3 5) reported in 

 19x1 cannot be criticised on the grounds 

 of failure to control luminosity, since 

 this investigator used figures constant in 

 area and intensity, differing only in form. 

 The two forms to be discriminated were a 

 circle and an ellipse. At first an ellipse 

 was used, the ratio of whose semi-axes 

 was as 2.M. The discrimination was 

 apparently made. This was followed by 

 using a series of ellipses gradually approx- 

 imating the circle in form. The ellipse 

 whose semi-axes bore the ratio 9:8 proved 

 to be the limit at which differentiation 

 just failed. Returning to the flatter and 

 longer ellipse, and reestablishing the 

 discrimination for it, the threshold was 

 again approached and with the same 

 result. The indication is therefore for 

 an acuity not far inferior to that of man. 

 This conclusion is, however, not supported 

 by any other modern work from the 

 Pavlov laboratory so far as we know, and 

 is, furthermore, contrary to the bulk of 

 evidence from all other sources. 



Can the dog see colors or is his visual 

 world composed of only lighter and 

 darker shades? The general opinion of 

 animal psychologists, in the light of tests 

 made on various forms, is that no mammal 

 below the monkeys and apes is capable 

 of color discrimination. 



No study of color vision in the dog has 

 been made by means of the Yerkes- Watson 

 method described above. So far as the 

 writers are aware the latest investigation 

 in this field is that of Smith (44) pub- 

 lished in 191Z. Experimentation on 

 color-vision in dogs has been, almost 

 without exception, of the crudest sort. 

 Even the experimenter has not been 

 eliminated from the situation. Prismatic 

 colors have never been employed. Usu- 



