THE INTRINSIC SYSTEMS OF THE FOREBRAIN 



1335 



Q 

 1-15- 



cruj 



LlI_J 



<o' 



_l 



<x 



h-cr 

 < 



LxJ 

 CO 



T 1 1 1 1 1 r 



-NORMALS 



■ TEMPORALS /' \ 



FRONTALS y 



r" 



-^ 1 r 



No. CUES in SITUATION 



8 



10 II 



12 



FIG. I I . Graph of the average of the number of trials tu 

 criterion taken in the multiple object experiment by each of 

 the groups in each of the situations after search was completed, 

 i.e. after the first correct response. (.See legends to figs. 7 and 

 1 o. ) Note the difference between the curves for the controls 

 and for the frontally operated group, a difference which is 

 significant at the .05 level by an analysis of variance (F = 8. 19 

 for 2 and 6 df), according to McNemar's (85) procedure 

 performed on normalized (by square root transformation) raw 

 scores. 



subsequent choices to the extent that 'finding the 

 peanut' determines subsequent choices for the normal 

 group. The experimental behaviorist, using terms 

 identical to those used by Sherrington in his lectures 

 on 'the integrative action of the nervous system,' 

 would describe the finding in more technical language : 

 for the group with frontal lesions, response to the 

 'positive element," i.e. the object with the peanut 

 under it, is inadequately 'reinforced' by the finding 

 of the peanut; as a result, the monkeys with frontal 

 lesions do not shift their responses to the reinforced 

 object as readily as do the controls. To state this more 

 generally, when given a choice, the intentions of 

 animals with frontal lesions are guided less than those 

 of controls by the behaviorally rele\ant consequences, 

 or 'outcomes,' of their prior actions. 



Interestingly, before the frontally operated group 

 begins to attain the necessary strategy (after the seven 

 cue situation), performance of tliis group reflects the 

 number of alternatives in the situation. This finding 

 suggests a parallel with analyses of the determinants 

 of intentions dexeloped in the theory of games and 

 economic behavior (141). Intentions are determined 

 by two classes of variables: a) the dispositions of the 

 organism and h) an estimate about the actions of other 

 parts of the system. The finding that performance of 

 the frontally operated group is related to the number 

 of alternati\es in the situation suggests that this group 

 is deficient in evaluating the second class of variables 



— but this is only suggested b\- these results. Support 

 for the hypothesis that frontal lesions do not affect the 

 dispositional variables that determine the preferences 

 comes from the results of another experiment. 



In a constant (fixed) interval experiment, 10 rhesus 

 monkeys are tested in an 'operant conditioning' ( 1 30) 

 situation which consists of an enclosure (discarded 

 icebox) in which a lever is available to the monkey. 

 Occasionally, immediately after a depression of a 

 lever, a pellet of food also becomes available to the 

 monkey. The experimenter schedules the occasions on 

 which the action of pressing the lever will make a 

 food pellet become available. In this experiment, these 

 occasions recurred regularly at a constant (fixed) 



40 

 c/,30 



LxJ 



o 

 85 10 



LlI 



o 40- 

 £ 30 - 



LlJ 



"20h 



UJ 



Q- 10 







~i I I 1 r 



PREHUNGER 



HUNGER 



POSTHUNGER 



70 HRS. DEPRIVATION -f 

 713- 1015 



18 HRS. DEPRIVATION 

 809- 1324 



X 



I 2 3 4 5 6 



20 sec. PERIODS 

 OF INTERVAL 



FIG. 12. Graph showing the effect of food deprivation on 

 monkeys" rate of lever -pressing response to food (a small 

 pellet of laboratory chow) which became available every 2 

 min. The change in total rate is indicated by numbers under 

 the deprivation label. The lack of change in the distribution 

 of responses is shown by the turves. Each curve represents 

 the average of the responses of 10 monkeys; each point repre- 

 sents the average rate during a period of the interval over 

 10 hr. of testing. \'ariance is indicated by the short honzintal 

 burs. (Dr. Nathan .^zrin made this experiment possible by 

 constructing apparatus and by suggesting that separate 

 coimters be used to record performance during each period 

 of the interval. Mr. David Nowel, Mr. Thomas Tighe and 

 Miss Libby Fleisher helped carry out this and the experiinent 

 reported in fig. 13.) 



