48 MILDRED A. HOGE AND RUTH J. STOCKING 



tinuecl. Neither of the rats in group 3 (rewarded) learned the 

 problem in 590 trials, at which time work had to be discon- 

 tinued with them also. The table shows also that the rats 

 exhibited great instability even until the end of the experi- 

 ment. A comparison of the errors of rats 3 and 7 shows that 

 rat 3 had never made a wholly perfect series of records before 

 her 520th trial, whereas rat 7 had made several perfect daily 

 records before the 30 errorless choices were made in succession. 



Fig. 2 shows the error curves for the three groups. The 

 ordinates represent the percentage of error; the abscissae the 

 number of groups of 10 trials each. It will be seen from these 

 curves that group 1, which was both punished and rewarded, 

 was the only group in which both rats made the discrimina- 

 tion. The curve of the punished group approaches that of 

 group 1 more nearly than does the curve of the rewarded group, 

 since one individual of the punished group learned the problem. 



Several interesting observations were made during the work. 

 The rats of the several groups maintained different attitudes 

 toward the work. Those animals which received the electric 

 shock showed after a few trials great hesitation before enter- 

 ing B, whereas the rats of group 3, which were never shocked, 

 rushed into B without making any apparent choice. The rats 

 which were punished became unwilling after a time to enter 

 the electric box, and they assiduously avoided the door to 

 this compartment, apparently associating the light with the 

 shock. Great patience was often required of the experimenters, 

 as several minutes would sometimes elapse before the rat could 

 be persuaded to come to the door I. When at length it passed 

 its head through this door, it would look to right and left 

 several times before making a choice. This indecision would 

 as often precede an incorrect choice as a correct one. How- 

 ever, as the discrimination became established, the animal 

 ceased to fear to enter the electric compartment. It would 

 come to the. door I, glance only once to left and to right, and 

 then dart toward the correct light. Although the rats which 

 received punishment were at first unwilling to enter the electric 

 compartment, they never became sulky as did the turtles used 

 by Casteel. 1 The rats remained in good condition throughout 



1 D. B. Casteel. The discriminative ability of the painted turtle. Jour, of 

 Animiil Behavior, 1911, vol. 1, no. 1. 



