SERIAL ADAPTATION 17/4 



pendent parts of four variables each or in one of eight. During 

 the random changes of trial and error, a field stabilising one of the 

 sets of four will occur many times more frequently than will a 

 field stabilising all eight (S. 20/12). Such a four, once stabilised, 

 will retain its field leaving only the other four to find a stable field. 

 Consequently, before the process starts we can predict that the 

 eight- variable system is much more likely to arrive at stability by 

 a sequence of four and four than by a simultaneous eight. The 

 fast process is the more probable (S. 16/7). 



We can predict, therefore, that in general if a multistable 

 system adapts to an environment composed of P independent 

 parts it will tend to develop P independent subsystems, each 

 reacting to one part. The nervous system, if multistable, will 

 thus tend to adapt to a fragmented environment by a fragmented 

 set of reactions, each complete in itself and having no relation to 

 the other reactions. It will do this, not because this way is the 

 best but because it must. But even though unavoidable, the 

 method is by no means unsuitable. It has the great advantage 

 of speed — it reduces to a minimum the dangerous period of 

 error-making — and there is no point in the nervous system's 

 attempting to integrate the reactions when no integration is 

 required. 



17/4. The second degree of complexity occurs when the environ- 

 ment is neither divided into independent parts nor united 

 into a whole, but is divided into parts that can be adapted to 

 individually provided that they are taken in a suitable order and 

 that the earlier adaptations are used to promote adaptation later. 

 Such environments are of common occurrence. A puppy can 

 learn how to catch rabbits only after it has learned how to run : 

 the environment does not allow the two reactions to be learned in 

 the opposite order. A great deal of learning occurs in this way. 

 Mathematics, for instance, though too vast and intricate for one 

 all-comprehending flash, can be mastered by stages. The stages 

 have a natural articulation which must be respected if mastery is 

 to be achieved. Thus, the learner can proceed in the order 

 ' Addition, long multiplication, . . . ' but not in the order ' Long 

 multiplication, addition, . . . ' Our present knowledge of mathe- 

 matics has in fact been reached only because the subject contains 

 such stage-by-stage routes. 



185 



