622 BRAIN MECHANISMS AND LEARNING 



Naquet. I think wc must be very careful about the site where the stimulation 

 of the reticular formation is effective. In this respect, I would like to mention some 

 experiments performed by Dumont, Paillas and Hugehn (1959), and by Hugelin, 

 Dumont and Paillas (1959), on the auditory pathways. These authors demonstrated 

 that if an inhibition of evoked response recorded at the level of nucleus cochlearis 

 exists in the animal 'encephale isole', this inhibition is not a consequence of a direct 

 action at the level of the first relay, but the result of muscle contraction of the 

 middle ear, induced by this stimulation. Finally, they show that if those auditory 

 muscles are either cut off, or curarized, the response at the level of nucleus coch- 

 learis remains constant in amplitude, but at the level of the auditory cortex it is 

 variable according to the state of synchronization or desynchronization of the 

 cortex. For these authors, this last finding would be due to a purely cortical 

 phenomenon; in our laboratory, with Drs Regis, Fernandez-Guardiola and 

 Fischer-Williams : if you put a cat under flaxedil, its pupils change with blood 

 pressure and with the state of synchronization and desynchronization of the cortex 

 - as was shown by Bonvallet, Dell and Hiekel (i954)- When the cortex presented 

 some spindles, the pupils narrowed and the response in the optic tract or in corpus 

 geniculatum laterale diminished in amplitude, simultaneously, at the level of the 

 cortex, the response was enhanced and the negative waves became more ample and 

 large and could be followed by a true slow after-discharge, having sometimes the 

 aspect of spindles. When the cortex was desynchronized (activated cortex), the 

 pupils were wide and the response in optic tract or corpus geniculatum laterale 

 increased in amplitude. Simultaneously, on the cortex, the amplitude of the 

 response diminished and its morphology was altered. The positive phase remained 

 unchanged and decreased discreetly in amplitude. The slow negative wave was 

 nearly totally disappearing. After atropine application, pupils being in constant 

 mydriasis had no reaction; there is no relation between their diameter and the 

 state of relative synchronization and desynchronization of the cortex. Evoked 

 potentials are very large at the level of optic tract and corpus geniculatum laterale 

 and do not vary, whatever the cortical clectrographic activity. At the level of the 

 cortex, the response followed the same variations as before application of 

 atropine. 



These data, as well as the results of Hugehn and co-workers, tend to prove that a 

 control of specific afFerences exists after the thalamic relay (as was suggested by 

 King et ai, 1956) and may play a more important role than in the first relay. 



JouvET. Investigations in man must be made with careful controls — with 

 atropinizcd eyes and recording of eye movement because even a slight movement 

 will interfere with the evoked potential at the cortical level. 



It is necessary to put some atropine in the eyes and at least record movements of 

 the eyes and eyelids by the electrogram of the blink and by the slow potential of 

 the eyes. This is absolutely necessary because with a flash that is more than twenty 

 centimetres from the eyeballs even a very slight eye movement would give a 

 reduction of the visual evoked potentials. 



Berger. What effect does the frequency of the stimulus have? Regarding 

 habituation, we have made experiments on what we have called fatigue. We 

 exposed the eye to flashes when we thought that attention was continually fixed on 

 the same subject. Depending on the area of the retina which was stimulated, the 

 general fusion frequency decreased significantly during a period of between 20 and 



