GENERAL DISCUSSION 647 



plays some role in determining whether given anatomical regions become posi- 

 tively or negatively rewarding. My second question has to do with an observation 

 that has been made repeatedly tor many years in psychological studies of learning. 

 Most of us have observed that learning curves do not really flatten out aiter the 

 initial phase ot acquisition but actually continue to demonstrate a considerable 

 amount of oscillation. In our experience this is true oi habituation as well as ot 

 positive learning. One wonders whether at least some portion of this oscillation is 

 not attributable to inborn excitability cycles which might determine the degree oi 

 responsiveness possible at any given moment in time. Biological rhythms of this 

 sort may have a wide variation in frequency ranging from months or days (circa- 

 dian rhythms) to cycles of very much shorter duration. Both Dr Hernandez-Peon 

 and I have been impressed with such oscillatory systems in our studies of habitua- 

 tion. 



Hebb. It seems to me that the meaning ot the selt-stiimilation data is exactly 

 opposed to that ot spontaneous behaviour. The seeking tor stinnilation in these 

 experiments is initiated by stimulation itselt, as the learning process goes on. The 

 nervous system acts in such a way as to produce an equivalent of external stimula- 

 tion: the action being by way of behaviour, bv bar pressing. 



Olds. I am not sure that I would come to exactly the same conclusion as l)r 

 Hebb. I think particularly relevant here is the case where we were reinforcing unit 

 discharges rather than reinforcing the overt response. In either case, though, let me 

 point out, that the response had to be omitted first, in order to be reinforced. We 

 really have in a sense two problems here ot the spontaneous or the innate, first that 

 the original response has to be in the repertory and has to occur by chance or 

 'reflex' before it can be reinforced; second that in the structure ot our organism 

 there has to be a hypothalamic mechanism or a system ot hypothalamic mechan- 

 isms which, when stimulated, will either cause the previous behaviour to be 

 repeated or, in one way or another, cause the animal to come back for more. That 

 is to say two things have to be here prior to learning: one is the spontaneous 

 response and the second is the h\pothalamic reinforcement mechanism whatever 

 we postulate it to be. 



One of the other points that I would like to put on record: The hypothalamic 

 reinforcement system appears to be an innately diflerentiated mechanism; i.e. 

 anatomically divided into functional groups. As an example I would cite the 

 precise anatomical determination ot self-stimulation rates. We are sure, in the rat, 

 that by fir the greater part ot the variance in self-stimulation rate is accounted for 

 by the placement of the electrodes. 



Let me now go on with the concept of reinforcement here. The behaviour occurs 

 spontaneously and there excitation in the hypothalamus occurs; at this point 

 something has to happen to the previous behaviour, or shall we say to the structure 

 in the organisms which somehow produces that behaviour or elicits that behaviour 

 — it must somehow become stronger, more frequent or what have you. I am greatlv 

 tempted, and perhaps it is solely a matter oi preference from these recent studies on 



