i486 



II WDBOOK HI- 1'IIVSIciI (x;y 



NEUROPHYSIOLOGY III 



least, neither the peripheral sensory nor the peripheral 

 motor response is required for conditioning. 



Self-Stimulation 



An entirely new line of brain stimulation experi- 

 ments began with the discovery (179) that under 

 proper conditions animals will work hard and con- 

 sistently to shock themselves if given the opportunity 

 to do so. In these studies a switch is put within reach 

 of an animal; when the switch is closed a train of 

 shocks is delivered through electrodes implanted into 

 certain regions of its brain. Monkeys, rats and cats in 

 this situation very soon spend most of their time clos- 

 ing the switch. It is clear that the brain shock serves 

 to reinforce the behavior, surprising as this may seem. 



The learning of this behavior may be extremely 

 rapid, for a very few brain shocks often suffice to make 

 the response rate reach the maximum which the ani- 

 mal ever displays. Once learned, the behavior is stable 

 and obeys many of the standard laws (219). There are 

 some indications that extinction, however, is very 

 quick, at least in the early stages of learning. Pre- 

 liminary reports of competition between this behavior 

 and other activities including learning have appeared 

 (42, 177, 178). 



It was found very early that shocks to all brain lo- 

 cations are not equally reinforcing and that, in fact, 

 the animal will not close the switch a second time for 

 shocks in some locations. Animals stimulate them- 

 selves actively when the electrodes arc in the limbic 

 system, or in mid-line structures in the hypothalamus 

 and rostral midbrain (17b). Negative locations are 

 closelv adjacent. 



It is still too early to assess the full significance of 

 these experiments. There is much interest at the 

 speculative level (which perhaps the reader shares) 

 regarding what the animal "feels" during self-stimula- 

 tion; perhaps direct tests will settle this point in man. 

 More objectively, the technique provides a clean new 

 method for classifying brain nuclei and tracts in terms 

 of whethei shocks to them are rewarding (i.e. rein- 

 force the behavior), punishing (i.e. depress it) or 

 indifferent. The preliminar) brain classification pres- 

 entl\ available, when compared with the ones derived 

 from anatomical and physiological techniques, has 

 ahead) revealed thai the punishing and rewarding 

 points Hud to lie in structures implicated in the moti- 

 vational, eiiioiion.il and attentive mechanisms in 

 learning. Ibis growing bod) oi information must 

 therefon be watched with much interest. 



Ihnni Shocks Influenci Learned Behavior 



When shocks arc applied to the unanesthetized 

 human brain exposed at operation the patient may 

 recount 'memories,' some of which could have been 

 deliberately learned. Presumably a particular set of 

 neural connections has been activated by the punctu- 

 ate stimulation applied (184). The barest beginning 

 has been made in attempts to apply this general idea 

 to animal experimentation and to produce changes in 

 acquisition and retention thereby. Shocks applied to 

 the cortex of rats improve learning of a maze and ac- 

 celerate formation of visual discrimination, but these 

 effects are barely significant statistically (72, 73). 

 Thalamic shocks applied to cats pressing a bar for a 

 food reward reduce the rate and increase the irregu- 

 larity of the response (37). Stimulation of the caudate 

 nucleus abolishes an avoidance response in dogs, but 

 shocks to the pulvinar fail to do so until they are made 

 very strong (36,). Clearly the information on this 

 point does not warrant any generalizations .it this 

 time. 



Electrot onvulsiue Seizurt 1 



The clinical observation that human beings under- 

 going convulsive therapy tend both to forget recent 

 events and to undergo changes in their emotional be- 

 havior has led to a number of experimental attempts 

 to define the neural mechanisms involved. Animals, 

 like people, quickly recover from the violent convul- 

 sions and unconsciousness produced by electrical cur- 

 rents passing through the head, and they appear 

 grossly to be normal some minutes later. In many 

 respects they ma\ well be. but careful testing has 

 alrcadx revealed that electroconvulsive seizures (ECS) 

 have certain specific effects on learning. 



ECS, for example, impairs acquisition. Duncan (53) 

 demonstrated this by training rats in an avoidance 

 task, subjecting some of them to ECS promptly after 

 each trial while delaying it lor others. The animals 

 receiving K( IS ju sec. alter each trial acquired the re- 

 sponse verv poorly, while those for which ECS was 

 delaved 1 hi. were indistinguishable from the con- 

 trols The intermediate delays (40 and 80 sec, 4 and 

 ij min. 1 vielded the expected intermediate degrees ol 

 learning. Entirely similar results have been reported 

 for hamsters in a maze test (75) and for rats in .1 dis- 

 crimination problem (234) I he fact that heat nar- 

 cosis at various times alter the final learning trial 

 affects learning in goldfish in a manner Comparable 

 to that of ECS in rats suggests that acquisition takes 



time in all animal forms I ; 



