J. C. ECCLES 349 



GROUP DISCUSSION 



Hebb. Was I correct in understanding that the excessively used synapses were 

 the ones that showed the most post-tetanic potentiation? Isn't this just the opposite 

 of what one would expect? They arc the ones which have been exposed the most 

 to the PTP, in their use, and we would now expect them to react the least. 



EccLES. The figures I showed you did indicate something like that. But you must 

 remember that these are reflexes in ventral roots and therefore quite unreliable as 

 samples of relative synaptic potency. I would not place any reliance on these 

 results with reflexes as measures of PTPs. Comparisons of relative sizes of PTPs 

 should be made by excitatory post-synaptic potentials. 



KoNORSKi. The view put forward bv Dr Eccles is that while excessive use ot given 

 synaptic connections increases their efliciency, the disuse produces opposite effects, 

 namely regression of synaptic function. This would be the simplest physiological 

 model of memorizing and forgetting. I held the same view several years ago (cf. 

 Konorski, Conditioned Reflexes and Nenron Organization, Cambridge, 1948), in 

 spite of the fact that a great body of evidence seemed to contradict it. Indeed, we 

 know very well that firmly established conditioned reflexes, as well as our own 

 memory traces remain intact even after many years of 'disuse'. But I couldn't 

 conceive any other mechanism of the phenomenon of forgetting than the atrophy 

 ot long disused synaptic connections. 



The realization of the great role played by dynamic memory traces in establish- 

 ing conditioned connections allows me now to overcome this difficulty and to 

 interpret the phenomenon of forgetting from quite a different point of view. As I 

 pointed out in my paper the chief feature of recent memory traces, as contrasted 

 with old memor)- traces, is that the)' are based on some dynamic mechanism 

 consisting probably in the transient activation of reverberating chains of neurones. 

 The ultimate fate ot these traces depends on the effects they produce : either they 

 give rise to appropriate morphological connections and are then transformed into 

 stable memory traces, or in certain conditions, as shown in my paper, they have no 

 chance to do so and are then totally obliterated. In such a case we have to do with 

 the phenomenon of forgetting. 



According to these considerations one is inclined to suppose that synaptic con- 

 nections once formed, whether in ontogenesis or as a result of special training, do 

 not atrophy by disuse, and that forgetting would probably concern only dynamic, 

 but not stable, memory traces. 



The recent results obtained by Dr Eccles seem to confirm my assumption. 

 Excessive use of the given spinal pathways led in his experiments to the increase of 

 efficacy of the synaptic connections involved, a fact analogous to the formation of 

 stable memory traces. But the disuse o{ the given pathways did not lead to the 

 deterioration of the reflex because of the stability of synaptic connections. 



Eccles. Now, of course, you cannot know what is going on in the dog's brain ! 

 He may not have been subjected to the experimental training procedures for long 

 periods, but he may be 'reliving' them in some experiential way. I realize that the 

 higher level synapses are probably quantitatively different from the lower ones, but 

 we can show qualitatively that the lower ones have the kind of properties you 

 would postulate as the basis of memor\- in the higher ones. I would not like to 

 comment on these disuse experiments until we have done a further series of 



