Brain Mechanisms

The neural connections of the ventromedial nucleus of the hypothalamus are consistent with what one would expect in a consociate modulator. Neurons of the ventromedial nucleus receive convergent inputs from the amygdala, septum, and midbrain tegmentum, including the midbrain central gray (Dreifuss & Murphy 1968; Tsubokawa & Sutin 1963), as well as anatomically-demonstrated inputs from the anterior hypothalamus (Chi 1970). The amygdala, in particular, is a brain structure that one would expect to be involved in processing the olfactory stimuli that make possible the recognition of consociates (Gloor 1978). The septum may be involved in discriminating conspecific opponents in terms of their vibrissal "jitter" rates, since both septal lesions and abolition of vibrissal jitter in the opponent lead to biting by rats in response to shock (Kanki & Adams 1978). Inputs from the amygdala and septum not only facilitate ventromedial neurons, but they also produce a complex excitatory-inhibitory effect thought to reflect complex feedback and feed-forward inhibition (Murphy 1972; Ono & Oomura 1975), such as one might expect from an "and-gate" logical device. There is also a major projection from the ventromedial nucleus to the midbrain central gray (Saper et al. 1976; Morrell et al. 1978); this corresponds to what one would expect if a ventromedial consociate modulator were the main input to a submission motivational mechanism in the central gray.

The outlines of a hypothetical neural circuitry controlling submission are shown in Figure 2. Afferent pathways from the amygdala, septum, and anterior hypothalamus, activated by stimuli from familiar consociates, converge upon a consociate modulator located in the ventromedial hypothalamus, which then sends projections activating the submission motivational mechanism in the midbrain central gray. At some point there is a logical and-gate that operates only when both inputs to the consociate modulator and motivating stimuli for defense and submission are simultaneously activated. The exact place where these inputs converge is not known; perhaps it is at the level of the ventromedial hypothalamus, or perhaps, as shown in the figure, it is at the level of the central gray. If the convergence occurs at the ventromedial nucleus, then there would have to be projections from the central gray to the ventromedial hypothalamus. There are some data indicating central gray projections to the ventromedial hypothalamus on anatomical (Szentagothai et al. 1962) and physiological (Beyer et al. 1962; Tsubokawa & Sutin 1963) grounds, but other anatomical studies have instead found that the central gray projections go only as far as the posterior hypothalamus (Chi 1970; Hamilton & Skultety 1970).

If one assumes that there are separate, parallel pathways for submission and defense in the forebrain, then it is possible to explain many peculiarities in the results of electrical stimulation. In some loci it is possible to obtain the full range of defense patterns by electrical stimulation; these include freezing, fleeing, and lunge-and-bite attack. In other loci it is possible to obtain freezing and fleeing, but without any accompanying affective defense or lunge-and-bite attack. Stimulation at the former loci, we may propose, activates the forebrain pathways that convey motivating cues for defense and submission. Stimulation at the latter loci, however, may activate the consociate modulator, which facilitates submission but not defense. In the cat, fleeing without affective defense may be elicited from a so-called "flight" zone of the amygdala (Ursin & Kaada 1960), and from areas that surround the perifornical region of the hypothalamus where defense may be stimulated; these areas include the anterior hypothalamus, lateral hypothalamus, and ventromedial hypothalamus, and certain areas of the central gray of the midbrain (Hunsperger 1956; Adams 1968). In the rat, fleeing without lunge-and-bite attack may be elicited from comparable regions, including the anterior hypothalamus (Woodworth 1971), lateral hypothalamus (Panksepp 1971), ventromedial hypothalamus (Woodworth 1971), and midbrain central gray (Waldbillig 1975). In the rat, in which boxing may be considered as a submissive response, dopaminergic stimulation of the corticomedial amygdala (Rodgers et al. 1976) and cholinergic stimulation of the basolateral amygdala (Rodgers & Brown 1976) can increase boxing in response to footshock without facilitating the lunge-and-bite response of defense.

There is a report, from work on the rat, that if electrical stimulation is confined within the ventromedial hypothalamus, the effect is suppression of defense without fleeing (Veening 1975). This might be predicted in case there was no stimulation of the perifornical system, which conveys motivating stimuli for defense and submission. In cats, on the other hand, stimulation of the ventromedial nucleus may produce either defense or fleeing (submission), perhaps by stimulation of the dendrites of perifornical neurons that extend into the ventromedial nucleus (Millhouse 1969).

The existence of separate parallel pathways for submission and defense is also indicated by the results of experiments in which two points are stimulated simultaneously, or in which both electrical and chemical stimulation are used. Whereas electrical stimulation in the hypothalamus of the cat produced defense (called "attack" by the author), chemical stimulation with carbachol from the same electrode produced submission (called "fear" by the author); when the chemical-stimulation effect was blocked, the electrical-stimulation effect remained intact (Baxter 1967). In another experiment it was shown that simultaneous stimulation of two fleeing points or two affective defense points in the cat summated in their effects, whereas simultaneous stimulation of a fleeing point and a defense point canceled each other's effects (Brown et al. 1969b). Further arguments in support of separate systems of submission ("flight") and defense may be found in Kaada (1967).

(Section continued on next page)