320 K. S. LASHLEY 



were made is due, at least in part, to the state of the animal's 

 health. She suffered at this time from a pulmonary trouble 

 which constantly distracted her attention from the stimuli. 



Control experiments showed that the rat was not depending 

 upon brightness, sound, or position in making the discrimina- 

 tion and her behavior in the discrimination compartment offers 

 the best evidence that she was really attending to the visual 

 stimuli. During experiments i and 2 she had always hesitated 

 for a long time before venturing from the starting box to the 

 discrimination compartment, finally dashing through to the food 

 box as quickly as possible, avoiding the punishment grill by 

 long leaps. When a marked difference in the brightness of the 

 forms was introduced she began to pause in the discrimination 

 compartment, swaying back and forth between the two passages 

 before choosing. As her proficiency in discriminating differences 

 in brightness increased, the time spent in choosing diminished 

 until the reaction became almost instantaneous. With the 

 greater number of correct choices her attempts to avoid the 

 punishment grill were less marked. With the introduction of 

 the lines as stimuli the hesitation between the passages again 

 appeared, increasing and diminishing gradually after each 

 change in the stimuli. 



From the percentages of error it is evident that the animal 

 could distinguish between the narrower rectangles. Her be- 

 havior in the discrimination compartment indicates that the 

 discrimination was more accurate than the percentage records 

 would seem to show, since they include all the trials made 

 without regard to accidental distraction of attention. After the 

 first day's work with the 20 by 30 millimetre rectangles, when 18 

 out of 20 trials were correct, the form of the stimuli was changed 

 as a control. With the new forms there was 60 per cent of error, 

 showing that perception of form was involved in the former 

 reaction, but unfortunately this short interruption was suffi- 

 cient to break up the association and in the next 300 trials 

 there was 45 per cent of error with complete loss of attention. 



EXPERIMENT 7. DISCRIMINATION OF SIZE 



Apparatus: The discrimination box; two circles of 30 and 

 50 millimetres diameter respectively as stimuli. In the type of 

 apparatus used the illumination of the two circular openings 



