VISUAL PERCEPTION OF THE CHICK 97 



In similar tests on relative size difference, however, the 

 results are more positive. These incidental tests were as 

 easily made with sizes as with forms. After the chick had 

 been trained to react appropriately to the o 28 + — o 12 + 

 stimuli, it w^as confronted, without further training, with two 

 different circles the sizes of which bore the same relation to 

 each other as that of the former pair. This test was made 

 with two different pairs of circles. One pair was larger, the 

 other smaller, than the training pair of circles, but the relative 

 difference was a constant. 



The diameters of the circles used ino 28+ — o 12+ discrimi- 

 nation are respectively 6 and 4 cm. The standard is thus three 

 halves of the variable. A habit having been established with 

 the 6-4 circles, these stimuli were replaced, during five tests, 

 with 4-3 circles; during five more tests, with 9-6 circles. Nearly 

 all of the chicks mentioned in Chapter V were tested in this 

 manner. 



The 9 cm. circles used in the tests with the larger pair of 

 circles was cut from black card board because regulation plates 

 of that size were not available. The circular opening was 

 neatly made by boring with a bit through two thin strips of 

 board placed on each side of the card board. In the dark 

 room human observers could hardly tell that it was a substi- 

 tute. Since there was no training with this cardboard plate, 

 the substitution could have had no influence. 



Not all chicks responded similarly to these novel size rela- 

 tions, but evidently those which were best trained responded 

 most positively to the relative stimulus difference. With all 

 of the subjects tested, the total positive reactions to the larger 

 stimulus considerably exceeded the total choices of the smaller 

 circle. Chick 21 responded without an error to both the 

 9-6 and 4-3 situations. Several other chicks responded 

 perfectly at least to one of the novel pairs of stimuli. On the 

 whole, the 9-6 test resulted in more correct reactions than the 

 4-3 test relation. 



In brief, these tests with new pairs of circles having approx^ 

 imately proportional size differences indicate that a chick 

 trained to choose a 6 cm. circle and to reject a 4 cm. circle can 

 choose, without further training, the 4 cm. circle when presented 



