MICHEL JOUVET 477 



closer to the 'basic' frcqucncv lost ctficacy earlier. Finally, all 'test' tones were 

 inopcrant. These observations confirmed that specificity of habituation ot EEG 

 'arousal' is not absolute, involving a certain degree of 'generalization' (5-20 c.p.s. 

 in 200 c.p.s.). Secondly, the biased progress of habituation to 'test' tones indicated 

 that after having learned not to respond to one pitch, learning not to respond to 

 similar frequencies was facilitated. Obviously, the process determining habituation 

 of EEG 'arousal' to a given pitch affects also mechanisms subserving 'activation' to 

 other tones, in a manner related strictly to the frequency arrangement upon the 

 pitch continuum (Apelbaum, Silva, and Frick 1959) Apelbaum, Silva, Frick and 

 Segundo, /;; preparation). 



(B) Coiiditiotiiii^. Tones reinforced (during wakefulness) by various absolute 

 stimuli have been presented to naturally sleeping cats. The following conclusions 

 could be drawn. 



(i) The effectiveness of each sound in provoking arousal (EEG and behavioural) 

 depended on the absolute excitation currently associated with it: reinforcement 

 with centre median or mesencephalic reticular stimulation conferred marked 

 arousing potency (Segundo, Roig and Sommer-Smith, 1959); reinforcement with 

 painful subcutaneous stimulation, less so (Galeano, Roig, Segundo and Sommer- 

 Smith, 1959); reinforcement with cessation of painful subcutaneous stimulation or 

 with application of amvgdaloid or caudate stimulation, little or no efficiency 

 (Roig, Segundo, Sommer-Smith and Galeano, 1959; Segundo, Galeano, Sommer- 

 Smith and Roig, this voltiDic). 



(ii) Certain discriminatory capabilities subsist during natural sleep and enable the 

 animal to differentiate positive (usually reinforced) from negative (usually non- 

 reinforced) stimuli; awakening takes place on presentation of the former but does 

 not occur after the latter. A degree of generalization exists, however, and is greater 

 for the EEG 'arousal' response than for the behavioural reaction; namely, if 200 

 c.p.s. are reinforced (and 210, 225 and 250 are not reinforced) EEG 'arousal' will be 

 produced by 200, 210, 225 and 250 c.p.s. tones; behavioural arousal on the other 

 hand only bv 200 and 210 c.p.s. (Segundo, Roig and Sommer-Smith, 1959). 



(iii) Conditioned inhibitory stimuli tend to have 'hypnogenic' qualities. Fig. i 

 illustrates this 'depressant' influence of 'negative' excitations contrasting with the 

 arousing potencv of the 'positive' stimulus (tone) (Segundo, Roig and Sonnner- 

 Smith, 1959). 



Consequently, and confirming observations from everyday life, learned issues 

 participate in determination of awakening. 



Hernandez-Peon. Drs George Bach-y-Rita and H. Brust-Carmona in my 

 laboratory have done some experiments on mesencephalic cats recording from the 

 cochlear nucleus and they observed a significant reduction of the auditory poten- 

 tials by repeating the same acoustic stimulus which has similarities to the same kind 

 of habituation observed in the intact cat. We think that it is possible to obtain 

 habituation in the mesencephalic cat. Dr Jouvet showed desynchronized tracings 

 in the mesencephalic reticular formation both in the intact cat during deep sleep and 

 in the mesencephalic cat which he believes to be permanently awake. How does he 

 decide when the desynchronized tracing represents sleep and when wakefulness? 



Jouvet. We use many indexes of behavioural sleep, including the classical 

 posture of the cat, the aspect of the pupils, the nictitating membranes, the rhythm 

 of respiration, the heart rate, the drop of rectal temperature. But we think that the 



