INTERACTION BETWEEN ADAPTATIONS 18/1 



usual living quarters where they were used for no further experi- 

 ments but served simply as breeders. Small groups were tested 

 from time to time for retention of the habit. 



4 The bird was fed in the dimly-lighted experimental apparatus 

 in the absence of the key for several days, during which 

 emotional responses to the apparatus disappeared. On the day 

 of the test the bird was placed in the darkened box. The 

 translucent key was present but not lighted. No responses 

 were made. When the pattern was projected upon the key, 

 all four birds responded quickly and extensively. . . . This 

 bird struck the key within two seconds after presentation of 

 a visual pattern that it had not seen for four years, and at 

 the precise spot upon which differential reinforcement had 

 previously been based.' 



I assume, therefore, that the system's behaviour is locally 

 regular, in the sense of S. 2/14. 



How will responses be localised in such a system ? Under- 

 standing is easier if we first consider the distribution over a town 

 of the chimneys that ' smoke ' when the wind blows from a given 

 direction. The smoking or not of a particular chimney will be 

 locally determinate ; for a wind of a particular force and direc- 

 tion, striking the chimney's surroundings from a particular angle, 

 will regularly produce the same eddies, which will regularly deter- 

 mine the smoking or not of the chimney. But geographically the 

 smoking chimneys are distributed more or less at random ; for if 

 we mark a plan of the town with a black dot for every chimney that 

 smokes in a west wind, and a red dot for every one that smokes 

 in a north wind, and then examine the j^lan, we shall find the 

 black and red dots intermingled and scattered irregularly. The 

 phenomenon of ' smoking '. is thus localised in detail yet dis- 

 tributed geographically at random. 



Such is the 4 localisation ' shown by the multistable system. 

 We are thus led to expect that the cerebral cortex will show a 

 ' localisation ' of the following type. The events in the environ- 

 ment will provide a continuous stream of information which will 

 pour through the sense organs into the nervous system. The set 

 of variables activated at one moment will usually differ from 

 the set activated at a later moment ; for in this system there 

 is nothing to direct all the activities of one reaction into one set 

 of variables and all those of another reaction into another set. 

 On the contrary, the activity will spread and wander with as 



191 



