RONALD E. MYERS 485 



being lowered. The choice situation was then programmed for the next 

 response by the patterns being exchanged or not exchanged between the 

 doors according to a predetermined chance sequence; the appropriate 

 stimulus door being baited ; and the intervening panels again lifted allow- 

 ing access to the choice chamber. The cats were subjected, in general, to 

 40 such training experiences or 'trials' each day. After a period of such 

 training, usually from a few days to a few weeks according to the diffi- 

 culty of the response, the animals achieved the criterion of learning on a 

 given discrimination, choosing the 'correct' pattern 34 or more times in 

 40 trials. After the animals had learned the discriminations they were over- 

 trained usually up to 400 additional trials to stabilize performance. 



A successful or high level performance by the animal, assuming ade- 

 quate control for undesired, non-visual clues, may be interpreted to mean 

 that the animal can see, that he perceives a difference between the two 

 discriminanda, and that his response to the stimuli is in keeping with 

 antecedent experience or conditioning in the training situation. This latter 

 capacity of specifying objects in relation to prior experience may be 

 defined as (Gliosis from the Greek word meaning kiiowk'dge, and implies 

 contribution from intact memory mechanisms. The specification of 

 objects in terms o( gnosis must be set off from their specification in terms 

 o( dimensions, location or movement or in terms o( colour, brightness or the 

 like. Our primary interest in this paper will be with I'isncil gnosis. 



PROBLEM I. IS THERE A SYSTEM OF GNOSTIC INTERCOMMUNICATION BETWEEN 



THE BRAIN-HALVES? 



Nine chiasma-scctioned cats were used in this first experiment. Four 

 were taught only a single discrimination through one eye (see Table I). 

 Cat Mnun may be described as illustrative of this group. Mmm was taught 

 and overtrained on discrimination I-ab while using the right eye. Final 

 performance through this eye was 39 correct in 40 trials. Did this cat 

 know using the left eye the pattern discrimination he had been taught 

 using the right eye? First performance through the right eye was 38 

 correct in 40 trails. This initial performance through the untrained eye 

 compared very favourably with the final performance through the 

 trained eye. 



The remaining five of the chiasma-scctioned cats were taught two or 

 more separate discriminations, usually one through each eye. Cat Bgw 

 may be taken as generally illustrative of all these animals. Bgw was taught 

 pattern discrimination Ill-ab while using his left eye. After criterion of 



