Understanding the mechanisms of population regulation in small mammals has long been a fundamental topic in population ecology [1, 2]. Hypotheses addressing population fluctuations and regulation in small rodents are generally classified into two categories: extrinsic or intrinsic hypotheses. Extrinsic hypotheses emphasize the role of climate , predators  and food  in causing population fluctuations, while intrinsic hypotheses emphasize the role of density-dependent genetics , physiology  and behavior [8, 9] . Both extrinsic and intrinsic factors can jointly contribute to population regulation in small mammals.
Density-dependence is well recognized in studies of population regulation driven by various intrinsic factors [10, 11]. The genetic regulation hypothesis , the physiological regulation hypothesis  and the social behavioral regulation hypothesis  are widely used to explain density-dependence in the population regulation of small rodents. The genetic hypothesis suggests that population density exerts selective pressure on different genotypes, favoring highly aggressive and low reproductive animals in high density, and vice versa. The behavioral regulation hypothesis suggests that territory defense, social rank and aggressive behavior play key roles in regulating populations; animals with low aggressive behavior or social rank suffer low reproduction and high mortality due to lack of resources. The physiological regulation hypothesis suggests that high population density induces high aggression and social stress, which result in disorders of the endocrine system including increases in corticosterone (CORT) and glucose levels, decreases in growth or reproductive hormones, followed by a population crash or decline . It is notable that all the three core hypotheses include the role of density-dependent aggressive behavior in regulating population fluctuations. Under laboratory condition, it is found that high density induced aggression or fighting could alter oxytocin (OT) and vasopressin (AVP) expressions in brains of Brandt’s voles , however, how population density affects aggressive behavior (and then population growth) of animals via neurological pathways has not been investigated in field conditions.
OT and AVP play a significant role in social recognition, aggression, parental care, mating [13-15]. Many previous studies have documented that OT, AVP and their receptors (i.e. OTR and AVPR) regulated aggressive behavior in mammals. High levels of aggression behavior in hamsters were closely associated with high AVP expression [16-18] . A high level of OT expression significantly decreased aggressive behavior in female rats (Rattus norvegicus)[19, 20].
Stress can change expression of AVP a nd OT and their releases , and then regulate the aggressive behavior of animals. In general, AVP could increase the stress responses, whereas OT could decrease stress responses [22, 23]. Population density of animals often fluctuates greatly in the field under the influence of climate, food, predators or parasites. High density is often accompanied by an increase in crowding, unfamiliar encounters and shortages of food or shelter, which may act as social or physiological stressors in animals. Our recent study indicates that under laboratory condition, high density as an environmental stressor could decrease OT expression, but increase AVP expression of Brandt’s voles; fighting could decrease OT expression but increase AVP expression of voles. In another experiment, injection of OT in brains reduced aggression of Brandt’s voles, but injection OT and OTR antagonists increased aggression of Brandt’s voles . Injection of AVP in brains could increase aggression of several other rodent species [25-27]. Decrease of OT or increase of AVP could mediate stress and reproduction hormone [28-30].
Based on the current knowledge (Fig. 1), we hypothesize that variation in population density will be reflected in variation of AVP and OT in the brains of small rodents in field populations. High population density as a stressor would increase both crowding stress and aggressive stress due to the increased encounters with strangers, which would then increase AVP expression but decrease OT expression in brains of animals. The high density induced the change of OT and AVP, further increase aggression behaviors of animals, and again increased the social stress, which in return increased AVP expression but decreased OT expression. The high density induced reciprocal enhancement between OT/AVP and aggression behavior would promote the level of aggression and stressful neuro-peptides. High levels of aggression should negatively affect population growth rates due to increased mortality or reproductive failure caused by fighting or interference. High level AVP or low level OT would increase release of stressful hormone (CORT) or decrease of reproductive hormone (e.g. GnRH, FSH, LH), and then reduce the reproduction or survival of individuals.
Brandt’s voles (Lasiopodomys brandtii) occur in the steppe grasslands of Inner Mongolia, China and Mongolia. They are social small herbivores which show large variations of population density in different years. Both extrinsic and intrinsic factors would affect their population fluctuations[31-33]. Our previous study on Brandt’s voles in laboratory condition demonstrated that at high density, AVP expression increased, and OT expression decreased in association with the observed increase of aggression behavior. We now test whether such associations occur in field conditions. We test our prediction that high density will increase AVP expression but reduce OT expression, and increase aggressive behavior of Brandt’s voles in large semi-natural field enclosures.