Author (Your Name)

Matthew S. Zweig, Colby College

Date of Award


Document Type

Senior Scholars Paper (Open Access)


Colby College. Psychology Dept.


James M. Gillespie

Second Advisor

E. Parker Johnson

Third Advisor

George S. Maier


Traditionally, killing or aggressive behavior has been viewed as an operant performed to gain reinforcement such as food. There is, however, increasing evidence indicating that killing is reinforcing in itself. Myer and White (1965) and Van Herrel (1970) have shown that natural killing behavior can be used to reinforce the learning of discrimination. Roberts and Kiess (1964) implanted electrodes in the anterior hypothalamus of natural non-killing rats and elicited killing behavior. They showed that this electrically elicited killing could also be used to reinforce the learning of discrimination. Thompson (1963,1964) demonstrated that both Siamese fighting fish and fighting game cocks would perform an operant to see a nonspecific and give their species-typical aggressive display. Tellegren et. al. (1969) showed that fighting mice will choose the arm of a T-maze which contained a mouse which they could attack. The relationship between feeding and killing has been thought to be a very close one. However, there has been a controversy in the literature as to whether killing is controlled by feeding and therefore a sub-system of a feeding mechanism or whether it’s an independent system in itself. Karli (1956) showed that natural killers would not eat the prey they had killed and natural non-killers would starve to death rather than kill. Von Hemel and Meyer (1970) and DeSisto and Huston (1970) found that natural killing rats would kill up to 30 mice and frogs without ever having ~he opportunity to eat them. Whalen and Fehr (1964) and Paul et. a1. (1971) have shown that cyclic food deprivation could increase the frequency of killing in a group of rats. Heimstra (1965) was not able to induce killing with cyclic food deprivation. Roberts and Kiess (1964) showed that hungry cats, from whom they were electrically elicited killing, would immediately leave a bowl of food to kill a rat when electrical stimulation was turned on. Hutchinson and Renfrew (1966) were able to elicit both feeding and killing from the same electrode although elicitation of killing was always at a higher current level. King and Hoebel (1968), DeSisto and Huston (1971) never observed feeding in stimuli-bound killers during stimulation and never observed killing in stimuli us-bound feeders. Karli and Vergnes (1964) showed that when both feeding and killing were abolished by rostral to caudal bilateral lesions within the hypothalamus, killing recovered before feeding in all cases. Thus killing took place with no feeding. The focus of the present set of experiments was to further clarify the relationship between killing and feeding mechanisms in the hypothalamus. Male Long-Evans hooded rats were implanted with bipolar electrodes aimed at the posterior lateral hypothalamus. They were tested for S-bound behaviors and S-bound feeders and killers were used in the experiments. The subjects were then trained to bar press. They received stimulation for as long as the bar was held down. Then either food, live frog or no goal object was placed above the bar and the rat was allowed to bar press and engage in an S-bound behavior simultaneously. The average bar press duration was recorded for each rat. The rats were run both food satiated and food deprived. A preference test was then run. Two bars were mounted at the end of the box and either no goal object, food, or frog were placed above each bar in various combinations. The rat was placed in the box and allowed to press at either bar.

The results were quite clear. The live frog and thus an opportunity to kill elicited longer durations of bar press for the S-bound killers than did the food or no goal object. For the S-bound feeders, the food elicited the longer duration. The results were unaffected by food deprivation. Thus, the effects of feeding and willing were differentiated in this experiment. The one S-bound killer run in the preference test showed a clear preference for the bar that had the frog over it.

These findings suggest that killing is in fact a mechanism in itself and one that is distinct from the feeding mechanism.


consummatory stimuli, brain stimulation reinforcement, hypothalamus, rats, feeding mechanism

Included in

Psychology Commons