7. An olfactory filter that is tuned to androgen-dependent pheromones and that facilitates the offense motivational mechanism and (possibly inhibits the exploration/marking motivational mechanism; organized perinatally and activated in adulthood by androgens.
In many muroid rodent species, there is a general tendency for adult reproductive males to attack other adult reproductive males. This is particularly well documented in the laboratory rat and mouse and its hormonal basis has often been reviewed (Bronson and Desjardins, 1971; Whalen, 1974; Leshner, 1975; Edwards and Rowe, 1975). The use of mice and rats in experiments has become so standardized that the encounter of an adult male mouse or rat with a strange male opponent has become known as "intermale aggression.' (Moyer, 1976). Since the attack is dependent upon an intact olfactory system (Ropartz, 1968; Bandler and Chi, 1972), and the presence of androgen in the opponent (Mugford and Nowel1, 1970; Lee and Brake, 1972; Thor and Flannelly, 1976), it may be supposed that it is activated by androgen-dependent pheromones.
The intermale aggression in other species of muroid rodents may be similar to that of rats and mice. The intermale fighting of golden hamsters and Mongolian gerbils is reduced by o1factory bulbectomy (Christenson et al, 1973; Hull et aI, 1974; Murphy, 1976) and depends upon the presence of androgen .in the opponent (Payne, 1974a). Female aggression, which also exists in these species, is presumably based upon other motivating stimuli such as the odors of an unfamiliar conspecific regardless of its sexual identity. In wild rats (Telle, 1966) and wild mice (Anderson, 1961), in which the odors of various individuals are presumably more heterogeneous than in the laboratory, there is fighting by females against opponents of both sexes.
The greater probability of offense by one male against another is dependent upon circulating androgen in all species in which it has been studied, including the laboratory rat (Barfield et al, 1972 mouse (Levy and King, 1953), hamster (Vandenbergh, 1971; Grelk et a1, 1974), and gerbil (Lumia et al, 1975; Yahr et a1, 1977).
The following line of reasoning suggests that androgen affects offense only by means of activating a sensory filter for motivating stimuli, rather than by affecting the offense motivational mechanism. First, as mentioned above, females of many species, and of wild rats and mice, show offense against unfamiliar conspecifics. Second, females engage in offense in a competitive fighting situation as much or more than do males (Zook and Adams, 1975).
The same olfactory filter, tuned to androgen-dependent pheromones and activated by androgen, may also inhibit exploration/marking behaviors (figure 1, site 7). This may explain why male mice spend less time in the part of an arena which contains urine from other intact males (Jones and Nowell, 1973 ). This effect was shown to depend upon the presence of androgen in the donor of the urine which elicited the effect (Jones and Nowell, 1973 ). Similar data have been obtained from male rats which approach and sniff odors from intact males less than from castrated males (Brown, 1977) and from male hamsters which approach and sniff castrated males in preference to intact males (Landauer et a1, 1977). Other motor patterns of exploration/marking may be affected, as one would predict if it were a function of a motivating stimulus to the motivational mechanism; thus male rats show significantly less urine-marking on areas which contain the odors of intact males than on areas with the odors of castrates (Brown, 1977).
There is some evidence that the site of the olfactory filter is in the forebrain in a region near the septum or preoptic area. Implantation of androgen in the septum of castrated male mice (Owen et al, 1974) or in the preoptic area of castrated male rats (Bermond, 1978) are able to reinstitute the intermale fighting that has been abolished by castration. Since these two areas are located close to each other, and since, in each study it seemed likely that there was some diffusion of the implanted androgen, it's not possible to say exactly where the effect took place. One can say, however, that the filter is probably not located in the amygdala, since lesions of the amygdala do not alter interma1e fighting in the rat (Busch and Barfield, 1974; Bunnell, 1966).
The olfactory filter that is tuned to androgen-dependent pheromones is organized perinatally by androgen. In the laboratory mouse, females which do not usually attack males can be made to show intermale fighting similar to that of males if they are administered androgen, around the time of birth (Edwards, 1969), and males that have been neonatally castrated show less fighting even if given replacement androgen in adulthood (Peters et aI, 1972). Similar data have also been obtained for the laboratory rat (Barr et al, 1916).