2/13 



DESIGN FOR A BRAIN 



correspondence is maintained exactly no matter how numerous 

 the variables. 



2/13. A system's field is the phase-space containing all the lines 

 of behaviour found by releasing the system from all possible initial 

 states. 



In practice, of course, the experimenter would test only a repre- 

 sentative sample of the initial states. Some of them will probably 

 be tested repeatedly, for the experimenter will usually want to 



15- 



10 



c 

 _o 

 O 



CP 



O 



£ 



5- 





10 15 



Weight of dog (kg. 



20 



Figure 2/13/1 : Arrow-heads show the direction of movement of the 

 representative point ; cross-lines show the positions of the representa- 

 tive point at weekly intervals. 



make sure that the system is giving reproducible lines of behaviour. 

 Thus in one experiment, in which dogs had been severely bled 

 and then placed on a standard diet, their body- weight x and the 

 concentration y of haemoglobin in their blood were recorded at 

 weekly intervals. This two-variable system, tested from four 

 initial states by four primary operations, gave the field shown in 

 Figure 2/13/1. Other examples occur frequently later. 



It will be noticed that a field is defined, in accordance with 

 S. 2/8, by reference exclusively to the observed values of the 



22 



