Updated Review-10

Many studies elicit attack behavior by electrical and chemical stimulation of the hypothalamus, but there are problems in interpreting the results. Behavioral results of stimulation are contradictory to the results of neural activity during similar naturally-occurring behavior. Neural activity in the hypothalamus has been found to be either unchanged in activity or inhibited during defense in cats evoked by the attack of another cat, even though electrical stimulation through the barrel of the electrode could often produce affective defense behavior (Adams, 1968a). Similarly, injections of carbachol in the hypothalamus of anesthetized cats decrease the firing rates of most neurons in the hypothalamic region (Brudzynski et al., 1991) where injection of carbachol produces affective defense in the awake animal (Brudzynski, 1981).

Furthermore, the nature of the attack behavior elicited in the hypothalamus by electrical and chemical stimulation is ambiguous. This has led one investigator who found it neither offensive nor defensive and varying from one strain of rat to another to call it a "behavioural category in its own right" (Kruk et al., 1990). Similarly, the attack obtained by Roeling et al. (1993) from bicuculline injection in the lateral hypothalamus appears to have been a mixture of offense and defense: ..."various forms of attacks like clinch fights, bite attacks and kick attacks [but] no sideways threat ". A more recent review (Kruk et al., 1998) provides a detailed analysis of the "equivocal outcome of such attempts to classify hypothalamic aggression into the motivational categories of offense and defense..."

One reason for the ambiguity of stimulation results may be the close proximity in the hypothalamus of cell bodies and axons involved in offense and defense motivational systems at several levels. 1) As described below, the hypothalamus apparently contains fibers of passage corresponding to projections from most motivating stimulus analyzers to the consociate modulator and defense motivational mechanism, and if it is indeed the case that the offense motivational mechanism is located in the anterior hypothalamus, projections to motor patterning mechanisms probably pass through the lateral hypothalamus where lesions abolish all the motor patterns of offense (Adams, 1971). 2) Neurons of the anterior hypothalamus, medial preoptic hypothalamus and ventral premammillary nucleus may be involved in the control of offense, while the dorsal premammillary nucleus has been implicated in the mediation of unconditional motivating stimuli of defense. 3) Hyperdefensiveness caused by lesions of the ventromedial nucleus of the hypothalamus led me to propose in the previous review (Adams, 1979a) that it may be the site of the hypothetical consociate modulator.

Another source of paradoxical results may be local inhibitory circuits such as those found in the central gray where single neurons are inhibited by nearby electrical stimulation (Sandner et al., 1986). With this in mind, it is possible that stimulation lateral to the ventromedial nucleus may produces defensive attack and flight by local inhibition, thus mimicking the effects of lesions of the ventromedial nucleus which produce hyperdefensiveness. This would be consistent with findings of Cheu and Siegel (1998) of reciprocal inhibition between lateral and medial hypothalamus. It would also be consistent with the finding that disinhibition of the ventromedial hypothalamus by local injection of bicuculline produces coordinated defensive flight in rats (Di Scala et al., 1984).

Because systemic pharmacological interventions may involve widely scattered brain structures and are therefore even more likely to have multiple effects on the various components of offense and defense, no attempt has been made to review this literature.