17/10 ANCILLARY REGULATIONS 



With this admitted, let us continue to examine those cases in 

 which some division into levels, and some easy understanding, are 

 possible. 



17/10. In Chapter 7 it was shown that the simple ultrastable 

 system would solve the basic problem of getting the primary 

 essential variables stable within their limits. But in Chapter 11 

 we recognised that adaptation, though it occurs in a purely logical 

 sense, may occur at such a low degree of efficiency as to be useless 

 for practical purposes. If we are to find the mechanism that 

 resembles the living, and especially the human, brain we must 

 find one that adapts, not merely in a nominal sense but with really 

 high efficiency. In S. 17/7 we found that such efficiency implies 

 adjustment of the degree of intra-cerebral connectivity to within 

 certain limits. 



We can now notice explicitly that there are other parameters 

 that will also have to be adjusted if the degree of adaptation is to 

 be more than merely nominal. Several of these have already 

 been noticed in passing: 



(1) In S. 8/15 we noticed that the duration of trial demanded 

 adjustment. In that section, the adjustment was, of course, made 

 by the operator before the tracings of the Homeostat's behaviour 

 were taken; but nothing has yet been said about how this adjust- 

 ment is to be made automatically in the organism. 



(2) In S. 7/7 it was demanded that the essential variables 

 should act on the step-mechanisms in the particular way: hunt 

 at ' bad ' and stick at ' good '. Nothing was said about how this 

 particular relation was to be provided in the organism. 



(3) In S. 10/8 it was shown that if an ultrastable system was 

 to adapt efficiently to a recurrent situation, a certain gating- 

 mechanism was necessary; but nothing was said about how the 

 organism should acquire one. 



(4) S. 13/11 showed how important is the value of the para- 

 meter : richness of equilibria among the states of the parts. 

 Nothing was said about how this parameter should be adjusted 

 to within satisfactory limits. 



Doubtless there are others that we have not yet noticed. One 

 other of outstanding importance deserves a section to itself. 



227 



